distributed energy storage with reactive and real...
TRANSCRIPT
DISTRIBUTED ENERGY STORAGE WITH REACTIVE AND REAL
POWER CONTROLLER FOR POWER QUALITY ISSUES BY
RENEWWABLE ENERGY AND ELECTRIC VEHICLES
By
LIM KHIM YAN
A dissertation submitted to the Department of Electrical and Electronic Engineering,
Lee Kong Chian Faculty of Engineering & Science,
Universiti Tunku Abdul Rahman,
in partial fulfilment of the requirements for the degree of
Master of Engineering Science
November 2016
ii
DECLARATION
I hereby declare that the dissertation is based on my original work except for
quotations and citations which have been duly acknowledged. I also declare that it
has not been previously or concurrently submitted for any other degree at UTAR or
other institutions.
Name : _________________________
Date : _________________________
iii
APPROVAL SHEET
This dissertation/thesis entitled “DISTRIBUTED ENERGY STORAGE WITH
REACTIVE AND REAL POWER CONTROLLER FOR POWER QUALITY
ISSUES BY RENEWWABLE ENERGY AND ELECTRIC VEHICLES” was
prepared by LIM KHIM YAN and submitted as partial fulfilment of the
requirements for the degree of Master of Engineering Science at Universiti Tunku
Abdul Rahman.
Approved by:
___________________________
(Prof. Dr. Lim Yun Seng)
Date:…………………..
Supervisor
Department of Electrical and Electronics Engineering
Faculty of Engineering and Science
Universiti Tunku Abdul Rahman
___________________________
(Mr. Chua Kein Huat)
Date:…………………..
Co-supervisor
Department of Electrical and Electronics Engineering
Faculty of Engineering and Science
Universiti Tunku Abdul Rahman
iv
The copyright of this report belongs to the author under the terms of the
copyright Act 1987 as qualified by Intellectual Property Policy of University Tunku
Abdul Rahman. Due acknowledgement shall always be made of the use of any
material contained in, or derived from, this report.
© 2016, Lim Khim Yan. All right reserved.
v
ACKNOWLEDGEMENTS
I would like to thank everyone who had contributed to the successful completion of
this research. I would like to express my gratitude to my research supervisor, Dr.
Lim Yun Seng for his valuable advice, guidance and his enormous patience
throughout the development of the research.
In addition, I would also like to express my gratitude to my loving parents
and friends who had helped and given me encouragement. Through this research, I
learnt a lot and I believe this helps me in my future. And I also appreciate helps and
advise from my co-supervisor Mr Chua Kein Huat, and Dr. Wong Jianhui in this
long journey.
vi
ABSTRACT
DISTRIBUTED ENERGY STORAGE WITH REACTIVE AND REAL
POWER CONTROLLER FOR POWER QUALITY ISSUES BY
RENEWABLE ENERGY AND ELECTRIC VEHICLES
LIM KHIM YAN
Substantial greenhouse gas emission due to the combustion of fossil fuels for
electricity has caused the world to push forward the use of renewable energy
sources. Photovoltaic systems and wind turbines are the two possible renewable
energy sources in Malaysia because of the convincing amount of solar irradiances
and wind energy throughout the year. Apart from that, electric vehicles (EV) have
been used to reduce air pollution in the cities. However, the integration of
renewable energy sources (RE) and electrical vehicles (EV) into the low voltage
(LV) networks can create a number of power quality issues, such as voltage
unbalance, low power factor and voltage rise issues when injecting power into the
networks. These power quality issues cause the circulation of the neutral current
around the distribution network, hence reducing the efficiency of the network.
Besides, the lifespan of induction motors can be shortened due to the unbalance
current around the motor windings making the motor to be overheated prematurely.
Hence, these power quality issues have to be mitigated. Therefore, an energy
storage system can be used to solve the power quality issues caused by the use of
RE and EV on the distribution networks. This is because the energy storage system
can supply or absorb the real and reactive power appropriately in order to mitigate
vii
the power quality issues. The advantage of using the energy storage system in this
aspect is that any power quality issues can be effectively corrected without wasting
the harvested energy. However, the operation of the energy storage system has to
be controlled appropriately in order to achieve the desired outcomes. Therefore, a
novel fuzzy controller is developed and implemented into the energy storage to
manipulate the flow of real and reactive power between the energy storage and the
network based on the network conditions. The fuzzy controller is able to reduce any
voltage excursions with the use of real and reactive power from the energy storage,
hence reducing the voltage unbalance and improving the power factor. Numerous
experiments have been carried out to verify the effectiveness of the fuzzy controller.
The experimental results showed that the voltage unbalances and power factors are
constantly maintained below the thresholds under the high penetration of PV
systems and EV.
viii
TABLE OF CONTENTS
Page
DECLARATION ii
APPROVAL SHEET iii
ACKNOWLEDGEMENTS v
ABSTRACT vi
TABLE OF CONTENTS viii
LIST OF TABLES xi
LIST OF FIGURES xii
CHAPTER
1.0 INTRODUCTION 1
1.1 Background 1
1.2 Aims and Objectives 4
1.3 Research Methodology 4
1.4 Research Outline 6
2.0 LITERATURE REVIEW 8
2.1 Embedded Generation 8
2.1.1 Photovoltaic System 9
2.1.2 Wind Energy 13
2.2 Electric Vehicles 16
2.3 Power Quality Issues Caused By the EV, PV and Wind Turbines 19
2.3.1 Voltage Unbalance Factor (VUF) 21
2.3.2 Voltage Flickers 25
2.3.3 Voltage Rises 26
2.3.4 Low Power Factor 27
2.4 Solutions for Mitigating the Power Quality Issues 31
2.4.1 Demand Side Management 31
2.4.2 Static Synchronous Compensator (STATCOM) 33
ix
2.4.3 Energy Storage System 34
2.5 Summary 37
3.0 METHODOLOGY Error! Bookmark not defined.
3.1 Introduction 38
3.2 Experimental LV Distribution Network 38
3.2.1 Network Emulator 41
3.2.2 Photovoltaic Systems 43
3.2.3 Wind Turbine Emulator 44
3.2.4 Electric Vehicles 47
3.2.5 Load Emulator 48
3.2.6 Energy Storage System 51
3.2.6.1 SMA Sunny Island 5048 52
3.2.6.2 Batteries 55
3.2.7 Communication and Data Acquisition Using LabVIEW 59
3.3 Summary 63
4.0 Design of Control Algorithm 64
4.1 Introduction 64
4.2 Fuzzy Controller for the Energy Storage System 64
4.2.1 Details of the Fuzzy Controller 68
4.2.2 Implementation of fuzzy Control Algorithm using
LabVIEW 73
4.2.2.1 Coding of the Data Acquisition 75
4.2.2.2 Saving of the data collected 76
4.2.2.3 Control of the Sunny Island 78
4.2.2.4 Coding of the Main Fuzzy Controller 79
4.2.2.5 Conditions for the Energy Storage System to
operate 82
4.3 Summary 84
x
5.0 RESULTS AND DISCUSSIONS 86
5.1 Introduction 86
5.2 Experimental Result and Discussion 86
5.2.1 Case Study 1: Impacts of high PV power output on the
experimental LV distribution network. 87
5.2.2 Case study 2: Power factor of a university’s building. 88
5.2.3 Case Study 3: Effect of using fuzzy controlled energy
storage on the network with PV. 90
5.2.4 Case Study 4: Effect of using energy storage when
inductive load is connected. 94
5.2.5 Case study 5: Effect of using energy storage when
electric car (EV) and photovoltaic system (PV) are
connected to the network. 96
5.2.6 Case Study 6: Impacts of wind power on the
experimental LV distribution network. 98
5.2.7 Case Study 7: Effect of using fuzzy controlled energy
storage on the network with wind emulator. 100
5.3 Summary 102
6.0 CONCLUSION 104
6.1 Conclusion 104
6.2 Future Works 105
PUBLICATIONS Error! Bookmark not defined.
REFERENCES Error! Bookmark not defined.
APPENDICES 120
xi
LIST OF TABLES
TABLE TITLE PAGE
2.1: Feed in tariff rates of the installed PV capacity (SEDA) 10
2.2: Average monthly wind speed for Magabang Telipot and
small Perhentian Island in 2005 15
2.3: Categories and Characteristics of Power System
Electromagnetic Phenomena 20
3.1: Specifications of VSD for network emulator 42
3.2: Specification of PV modules 43
3.3: Specifications of VSD for wind emulator 46
3.4: Specifications of SMA Windy Boy 2500 46
3.5: Specifications of SMA Sunny Island 5048 54
4.1: Rules of defuzzification for voltage change (ΔVy) 70
4.2: Rules of defuzzification for power factor change (ΔPFy) 71
5.1: Comparison of the improvement of VUF with the
integration of the energy storage system 93
5.2: Improvement of power qualities with the integration of
the fuzzy controlled energy storage system 95
xii
LIST OF FIGURES
FIGURE TITLE PAGE
2.1: Configuration of single-phase PV system 10
2.2: World map showing the average cloud amount in
percent 12
2.3: Fluctuation of voltage rise, voltage unbalance factor and
power factor due to intermittent PV output power 13
2.4: Generation of wind energy 14
2.5: Architecture of electric vehicle (Forster, 2010) 17
2.6: The locations of existing charging stations in Malaysia
(First Energy Networks, 2015) 18
2.7: Public charging stations located at P2 of Suria KLCC
(KeeGan Dorai, 2012) 18
2.8 (a): Phasor of the balanced three-phase voltage
(b): Phasor of the unbalanced three-phase voltage 21
2.9: Increase in motor heating due to voltage unbalance 23
2.10: Relationship between real, reactive and apparent power 27
2.11: Normal power factor in the network 299
2.12: Power factor being reduced when the PV system is
integrated 29
3.1: Experimental network diagram 39
3.2: Flexible LV panel 40
3.3: Experimental setup 40
3.4: Grid emulator 41
xiii
3.5: Variable speed drive used to drives the induction motor 42
3.6: Synchronous generator coupled to an induction motor 42
3.7: PV system 44
3.8: Photovoltaic panel mounted on a metal structure 44
3.9: Topology of the wind turbine emulation system 45
3.10: (a) Induction motor coupled with synchronous machine
driven by (b) VSD of wind turbine emulator 45
3.11: (a) Rectifier; (b) SMA Windy Boy 2500 46
3.12: Electric vehicle used in the experiment 47
3.13: Charging connector of the electric vehicle 48
3.14: Load bank used in the experimental 49
3.15: NI 9403 37 channels connection diagram 49
3.16: D2425 Solid State Relay 50
3.17: Arrangement of controllable load bank 50
3.18: Inductive load used in the experiment 51
3.19: Setup of the proposed energy storage system 52
3.20: Sunny Island 5048 bi-directional inverter 53
3.21: Batteries 57
3.22: Arrangement of energy storage system 57
3.23: Depth of discharge for C10 Hoppecke solar.bloc 58
xiv
3.24: CompactDAQ 9174 Data Acquisition Chassis 59
3.25: Data Taker DT82 Intelligent Data Logger 60
3.26: Data acquisition between PC, Data Taker, cDAQ 9174,
and the bi-directional inverter 60
3.27: Screenshot of LabVIEW program 62
4.1: Flowchart of control algorithm 67
4.2: Fuzzy logic input variable membership for voltage
change (ΔVy) 69
4.3: Fuzzy logic input variable membership for power factor
change (ΔPFy) 69
4.4: Fuzzy output membership of instruction (ΔIp) 71
4.5: Fuzzy output membership of instruction (ΔIq) 71
4.6: CoA defuzzification method for real current change (ΔIp) 73
4.7: CoA defuzzification method for reactive current change (ΔIq) 73
4.8: Graphical User Interface of the Energy Storage System 74
4.9: Load bank control 75
4.10: Values read from the input registers 75
4.11: Way to decode the encoded the parameters obtained 76
4.12: Data saved in excel format 77
4.13: Data Collection coding 77
4.14: Turning ON and OFF of the Sunny Island 78
4.15: Controlling of Sunny Island 79
4.16: Read values from Sunny Island 79
4.17: Calculation of VUF 80
4.18: Park’s Transformation 81
xv
4.19: Phase selection for low power factor 81
4.20: Fuzzy logic for voltage unbalance 82
4.21: Fuzzy logic for power factor 82
4.22: Condition of active current before sent to the energy
storage system 83
4.23: Processing of reactive current before sent to the energy
storage system 84
5.1: PV power output at Phase A, load at Phase A and power
factor of all three phases of the experimental case study 1 87
5.2: VUF and voltage at phase A of the experimental case
study 1 88
5.3: Power factor profile on 1st July 2013, Monday 89
5.4: Power factor on 6th July 2013, Saturday 900
5.5: Three-phase voltages and PV power output of the
experimental case study 3 91
5.6: VUF and power output of the energy storage system of
the experimental case study 3 92
5.7: Corresponding VUF of the energy storage system using
the control strategy of (Wong, Lim, & Morris, 2014) 92
5.8: Power factor of all three-phases and reactive power
output of the energy storage system 94
5.9: (a) Three-phase voltages, (b) VUF,
(c) Reactive power flow and, (d) Power factor of the
experimental case study 4 95
5.10: VUF, voltage, active power output of energy storage
system, and PV power of the experimental case study 5 97
5.11: Power factors of all three phases and reactive current
output from the energy storage system 97
5.12: Mean daily wind speed values at Kuala Terengganu 98
5.13: Wind power output at phase A and power factor of
three-phases in experimental case study 6 99
xvi
5.14: VUF and three-phase voltages of the experimental case
study 6 100
5.15: Three-phase voltages and wind power output of the
experimental case study 7 101
5.16: VUF and power output of the energy storage system of
the experimental case study 7 101
5.17: Power factor of all three-phases and reactive power
output of the energy storage system for the experimental
case study 7 102
CHAPTER 1
1 INTRODUCTION
1.1 Background
Renewable energy has become a very important energy source in Malaysia to deal
with the climate change and reduce the dependency on fossil fuels for energy.
Photovoltaic (PV) system is the most promising renewable energy resource in
Malaysia due to its abundant solar irradiance. However, most of the PV systems are
single-phase and installed on the low-voltage distribution networks without proper
planning. As a result, the voltage unbalance of the network may be increased,
causing a large circulation of neutral currents. Consequently, the network efficiency
is reduced because of the power losses from the high neutral current. In addition,
Malaysia is one of the cloudiest countries in the world. The solar irradiance is often
scattered due to the passing clouds. As a result, the PV power outputs tend to
fluctuate substantially. The intermittent power outputs and the growth of PV systems
on the LV network can cause the high voltage fluctuations and reverse power flow
(Prakash K. Ray, 2013).
2
Promoting electric vehicles is another approach put forward by the government to
create a clean environment. Therefore, electric vehicles are increasingly popular in
the country. There are 19 charging stations at present and another 300 charging
stations will be installed by year 2015. Therefore, power consumption can be
increased with the possibility of causing a huge burden to the electrical network.
Also, the use of the charging stations can increase the voltage drops and voltage
unbalance on the distribution networks; hence, causing the power factor drops. When
the power factor is low, significant heating effects and the premature failures of
transformers and electrical machines can happen. Also, customers can be penalised
by the utility companies if their power factors are lower than 0.85. Any voltage
instability can damage the electrical motors or result in the disoperation of electronic
equipment (Canada, 2004).
All the technical issues created by the renewable energy and electrical vehicles need
to be dealt effectively in order to accommodate the rapid growth of the renewable
energy and electrical vehicles in the future. There are a number of solutions available
to maintain the network voltage such as static VAR compensator, static synchronous
compensator and active filters (Li, Zhang, Tolbert, Kueck, & Rizy, 2008). These
devices control the network voltage with the use of reactive power. Therefore, these
devices are an effecitve on the transmission networks because the reactance is
usually large than the resistance of the network (von Appen, Marnay, Stadler,
Momber, Klapp, & Scheven, 2011). However, they may not be effective on the
distribution networks where the resistance is larger than the reactance.
3
To solve the technical problems on the distribution networks, it is proposed to use
distributed energy storage systems which can maintain the network voltage within
the required tolerance with the use of both real and reactive power. A fuzzy
controller has been developed to control the real and reactive power output of the
energy storage system in order to solve the voltage fluctuations on the distribution
network caused by the penetration of the single-phase renewable energy sources and
the electrical vehicles. This fuzzy controller is the extension of the method proposed
in (Wong, Lim, & Morris, 2014) which uses real power only to maintain the network
voltage. This extended fuzzy controller uses both the real and reactive power to
maintain the network voltage, voltage unbalance and the power factor of the
network, hence improving the efficiency of the network.
The fuzzy controller developed comprises the fuzzy logic control and the Park’s
transformation in order to deal with fluctuating voltage rises and voltage unbalances.
The Park’s transformation is used to coordinate distributed energy storage systems
on the three-phase network to balance the voltage effectively. The controller involves
the manipulation of the reactive power from the energy storage systems in order to
maintain the power factor. The performance of the fuzzy control algorithm is verified
by conducting a series of experimental case studies. The experimental results showed
that the fuzzy control algorithm is able to solve the voltage unbalance issue, maintain
the voltage magnitude at its statutory limit, and improve the power factor of the
network.
4
1.2 Aims and Objectives
The aims and objectives of this project are as follows:-
I. To customize a distribution network integrated with two PV systems,
a wind turbine emulator and an EV charging station.
II. To develop an energy storage system integrated with the distribution
networks.
III. To develop a fuzzy control algorithm for the energy storage system
for mitigating power quality issues using both the real and reactive
power from the energy storage system.
IV. To evaluate the performance of the fuzzy controller in mitigating the
power quality issues such as voltage unbalance, low power factor, and
voltage rise caused by PV and EV.
1.3 Research Methodology
In order to achieve the objectives of this project, the following research activities
were carried out.
Step 1: Literature study
Literature review was carried out to understand the incentives launched by the
government to promote the growth of photovoltaic systems and electrical vehicles on
the distribution networks. The embedded generation system in Malaysia with PV,
5
wind energy, and electric vehicles was also studied. In addition, power quality issues
caused by the integration of PVs, wind turbines, and EVs together with the solutions
for mitigating the issues are also studied.
Step 2: Development of LV network
An experimental network emulator integrated with photovoltaic systems and
electrical vehicle charging station was developed at the university campus. The
network consists of PV systems, wind turbine, wind emulator, load emulator, and
data acquisition equipment.
Step 3: Impact Studies
Various case studies were carried out to investigate the power quality issues caused
by the single-phase PV systems and the electrical vehicle on distribution network
emulator.
Step 4: Setup of energy storage system
Three-phase energy storage system was set up in the university’s laboratory. 48
batteries were connected to the three bi-directional inverters which were connected to
the grid.
Step 5: Simple control for the energy storage system
Communication and simple control technique was developed to control the real and
reactive power output of the energy storage system manually to improve power
quality issues.
6
Step 6: Fuzzy controlled energy storage system
Fuzzy control algorithm which uses real and reactive power of the energy storage
system was developed to mitigate the power quality issues.
Step 7: Evaluation of the fuzzy control algorithm
Few case studies were carried out with different generation and loading conditions
on the distribution network to assess the performance of the fuzzy controller in
mitigating power quality issues caused by the PV and EV on the distribution network.
Step 8: Implementation of the fuzzy control algorithm for the three-phase energy
storage system
Due to the non-uniformly installation of single-phase PV system, a fuzzy controlled
three-phase energy storage system was further developed from step 6. Experiments
were carried out to verify the performance of the fuzzy controlled three-phase energy
storage system.
1.4 Research Outline
The layout of the thesis is structured as follow:-
Chapter 2 presents the literature review on the distributed generation. The potential
grid connected renewable energies are briefly discussed, followed by the power
qualities arise due to the grid connected renewable energy on distribution networks.
7
Electric vehicles which cause various power quality issues are also explained,
followed by the literature review on solutions to mitigate power quality issues.
Chapter 3 describes how the experimental LV distribution network was constructed.
The design and the development of the electrical system, emulator, renewable
resources, communication and data acquisition are stated clearly. In this chapter, the
design and development of the energy storage system are briefly described. It is
followed by the description of the control strategy. Details of fuzzy control algorithm
are elaborated.
Chapter 5 presents several case studies conducted to represent different generation
and loading scenarios. Impacts caused by the penetration of renewable energy at the
university premises on distribution network were investigated. Different generations
and loading conditions were considered to evaluate the performance of the proposed
energy storage system to mitigate power quality issues.
Chapter 6 is the conclusion for the key findings of the research work and the
uniqueness of the proposed energy storage system.
8
CHAPTER 2
LITERATURE REVIEW
2.1 Embedded Generation
Dispersed generation or embedded generation refers to any power generation
technologies integrated with the distribution systems which is close to the points of
load demands. They are not centrally planned by the utility or not centrally
dispatched. Generally, they are smaller than 50 - 100 MW and connected to the
distribution system. Over the last few years, the popularity of embedded generation
is blooming due to few reasons. Based on the survey of International Conference on
Electricity Distribution (CIRED), the reasons for the growth of embedded generation
are the national drivers for reducing greenhouse emissions, improving energy
efficiency, improving diversification of energy resources and the security of power
supply. Moreover, for commercial considerations, distributed generation is available
in modular and it is smaller in space occupied. Embedded generation may reduce the
transmission cost and the power losses due to the long transmission of electricity
(Jenkins, 2000).
9
The significant penetration of embedded generation will cause reverse power flow,
voltage unbalance, low power factor, voltage flickering, voltage rise, voltage sag,
harmonics and etc. Embedded generation contains a wide range of technologies such
as solar power, wind turbine, combined heat and power (CHP), energy storage
system, biomass, hydropower, ocean renewable energy, and etc. Solar energy, wind
energy, hydropower, and fuel cell are the common renewable embedded generation
in Malaysia.
2.1.1 Photovoltaic System
Photovoltaic (PV) system consists of solar panels and inverters for converting the
sunlight into electricity. The PV inverters are to convert the solar electricity from
direct current (DC) to alternating current (AC) before the solar electricity is exported
to the grid. Figure 2.1 shows the configuration of the single-phase PV system. It can
be a single-phase or three-phase PV system. However, the single-phase PV system is
relatively common and widely installed in household buildings. The PV systems can
be ranged from small roof top mounted or the building integrated systems from few
kilowatts to megawatts. The PV systems are environmental friendly as they generate
little greenhouse gaseous throughout their life cycles. In addition, the operating and
maintenance fees are low because the PV systems are a mature technology with the
service life of 20 years or more.
10
Figure 2.1: Configuration of single-phase PV system
The 8th Malaysia Plan on Energy Policy promoted several incentives to popularize
clean energy. The PV system is highly encouraged as it generates clean energy from
sunlight. Moreover, feed-in-tariffs and funding on renewable energy were announced
in the 10th Malaysia Plan in 2010. Under the feed-in-tariff mechanism, installed PV
capacities granted with feed-in approval are approximately 192.80 MW (SEDA).
Table 2.1 shows the feed-in-tariff for the installed PV system. Costs are dropping due
to the mass production of the technology. PV will one day be cost-effective in urban
areas as well as remote ones. PV will be one of the most promising renewable energy
resources in Malaysia.
Table 2.1: Feed in tariff rates of the installed PV capacity (SEDA)
Basic FiT rates having installed capacity of: FiT Rates (RM per kWh)
(i) Up to and including 4 kW 0.9166
(ii) Above 4 kW and up to and including 24 kW 0.8942
(iii) Above 24 kW and up to and including 72 kW 0.7222
(iv) Above 72 kW and up to and including 1 MW 0.6977
(v) Above 1 MW and up to and including 10 MW 0.5472
(vi) Above 10 MW and up to and including 30 MW 0.4896
The integration of PV systems into the networks can cause power quality issues such
as voltage flickers, voltage fluctuations, voltage unbalance, transient voltages,
voltage rises, and harmonics due to the lack of coordination between customers and
11
utility companies, active power variations, and reactive power flow. In addition, the
network experiences the low power factor (R. Romo, 2015). Therefore, with the
penetration of PV, the machinery and motors can be damaged easily or the life span
of the equipment will be shortened due to the power quality issues in the electrical
networks.
Figure 2.2 is the world map showing the average amount of cloud in percentage
(Ryan Eastman, Climatic Atlas of Clouds Over Land and Ocean, 2014). It shows that
Malaysia has a high coverage of clouds. Hence, the cloudy sky of Malaysia is also an
issue because the power output of the PV systems can be highly intermittent. It is
hard to estimate the inconsistent power acquired due to the minute-to-minute
variations. Besides, these cloudy sky will also create fluctuating power quality issues.
Therefore, the power quality issues caused by the intermittent PV output power are
unique because they are fluctuating and difficult to be solved because most of the
existing technologies is to solve steady state voltage rise, voltage unbalance and
power factor. (Chua, et al., 2012)
12
Figure 2.2: World map showing the average cloud amount in percent
Figure 2.3 shows the data collected when a single-phase PV system is integrated with
the distribution network. It shows the fluctuating voltage, voltage unbalance factor
and power factor due to the intermittent PV output power. Hence, it is important to
mitigate the technical issues caused by PV systems in order to maintain the network
stability and reliability in the future. However, the utility company needs an
advanced active control in order to cater for these power quality issues.
13
Figure 2.3: Fluctuation of voltage rise, voltage unbalance factor and power factor due
to intermittent PV output power
2.1.2 Wind Energy
Wind turbines operate on a simple principle. The energy in the wind turns the blades
around the rotor connected to the shaft, hence driving the generator to generate
electricity. Therefore, the wind turbines convert the kinetic energy from wind into
useful electricity. A wind controller and AC to DC converter are used to generate the
electricity from the three-phase AC to the single-phase DC output which will be
converted to single-phase AC 240 V at 50 Hz. Figure 2.4 shows the generation of
wind power.
14
Figure 2.4: Generation of wind energy
In Malaysia, wind power is not popular compared to the solar power because the
wind speed is not consistent. The wind speed is low throughout the year, 2 m/s, on
average. Nevertheless, the research carried out at Serdang area showed that doubling
the height of the wind turbine can increase the wind speed by 10% (Teh, 2010).
Besides, harnessing wind energy in Malaysia depends on the geographical location
since Malaysia is situated in the equatorial region and the wind speed relies on the
monsoons. Furthermore, (S.Mekhilef, 2011) shows that the locations facing the
South China Sea are the best choices for off-shore wind farms as there is Northeast
monsoon season winds from November to February. Table 2.2 shows the average
monthly wind speed for Magabang Telipot (Zaharim, Najid, Razali, & Sopian, 2009)
and small Perhentian Island (Zuhairuse Md Darus, 2009) in 2005. It is shown that it
is possible to harness wind energy in Small Perhential Island because the cut-in
speed of small scaled wind turbine is 3 to 4 m/s.
15
Table 2.2: Average monthly wind speed for Magabang Telipot and small Perhentian
Island in 2005
Month Wind speed (m/s) for Megabang Telipot Wind speed (m/s) for Small Perhentian Island
January 4.44 7.9
February 3.47 6.85
March 3.16 7.56
April 2.54 6.78
May 2.24 6.91
June 2.25 nil
July 2.18 6.63
August 2.13 7.16
September 1.91 6.88
October 2.4 7.12
November 2.38 7.57
December 4.58 7.88
The integration of the wind turbines with the networks can cause power quality
issues such as voltage fluctuations, voltage rises, and voltage sags due to the
switching operation, low power factor, and harmonic distortions (Sharad.W.Mohod,
2008). In addition, wind does not blow uniformly throughout Malaysia. Hence,
power quality issues are fluctuating due to the intermittent wind power output.
Fluctuating power quality issues are hard to be mitigated since the existing strategy is
used to mitigate steady power quality issues. With the continued expansion of the
wind power generation, the fluctuating power quality issues are concerned widely.
For the existing solutions, the static synchronous compensator (STATCOM) is used
as a smooth reactive power provider to improve the power quality issues caused by
the wind integrated electrical networks (Muhammad A.Saqib, 2015). However, the
STATCOM is expensive and an innovative strategy should be developed to mitigate
the fluctuating power quality issues caused by the integrated wind power generation.
16
Therefore, an energy storage system with an effective controller is proposed to solve
the dynamic power quality issues caused by the wind turbines.
2.2 Electric Vehicles
Institute of Energy Research states that transportation has occupied 70% usage of
petroleum, which is the highest compared to other usages. Green technology is one
of the highly recommended ways to mitigate the issue while electric vehicle (EV) is
the new highlight in this era to overcome this problem. Based on the analysis from
GreenTech Malaysia, the migration to electric mobility from petrol vehicles will
potentially help Malaysia to cut down the carbon dioxide emissions by 1.7 tonnes in
the long run (Lim, Chia Ying, 2015). Electric vehicle consists of electric motor,
DC/AC inverter, DC/DC converter and batteries. The architecture of electric vehicle
is illustrated in Figure 2.5 (Forster, 2010). Since high voltage is required for the
electrical motor to perform, DC/AC inverter plays an important role to draw power
from the batteries. Batteries are controlled and monitored by the battery management
system of the electric vehicle in order to avoid over discharge.
17
Figure 2.5: Architecture of electric vehicle (Forster, 2010)
Electric vehicles are a step closer to creating the world a greener place. Widely usage
of electric vehicles indicates that charging stations will be installed everywhere. At
present, electric vehicles are getting popular among the country and it is available for
purchase in Malaysia. The number is expected to grow in the near future. In
Malaysia, 19 charging stations have been installed and another 300 charging stations
will be installed by year 2015. Figure 2.6 shows the map of the existing charging
stations in Malaysia. The available charging stations are installed at Suria Kuala
Lumpur City Centre (KLCC), Lot 10 Shopping Centre, Petronas Solaris Serdang,
Bangsar Shopping Centre, Wisma Tan Chong, Nissan Petaling Jaya, and etc (First
Energy Networks, 2015). Figure 2.7 shows the public charging stations located at P2
of Suria Kuala Lumpur City Centre (KLCC).
18
Figure 2.6: The locations of existing charging stations in Malaysia (First Energy
Networks, 2015)
Figure 2.7: Public charging stations located at P2 of Suria KLCC (KeeGan Dorai,
2012)
Power quality issues caused by the penetration of EV and PV are discussed in
(Masoud Farhoodnea, 2013). Simulation results showed that a severe voltage drops
due to the high power consumption and unmanaged arrangement of the EV charging
19
stations. Due to the unplanned connection to the networks, voltage unbalance is
likely to happen. Besides, charging EV with the electrical network will cause voltage
variations, power factor reduction, voltage drop and harmonics distortion. In the
paper, the authors also state that capacitor banks or active power conditioning
devices must be used to manage the exchanged powers and control the voltage of the
networks. There are some solutions proposed to mitigate these problems. In (Yong,
Ramachandaramurthy, Tan, & Mithulanantha, 2015), the authors use bi-directional
DC fast charging stations to solve the voltage drop issue. By controlling the
switching, reactive power drawn from the DC-link capacitor can improve the power
factor. Besides, vehicles-to-grid (V2G) with various charging strategies of EV as
shown in (Salman Habib, 2015) can improve the technical performance of the
networks in terms of stability, efficiency and reliability of the systems.
2.3 Power Quality Issues Caused By the EV, PV and Wind Turbines
The power quality has become an important issue in the distribution networks in the
recent years due to the grooming of power electronics used and the distributed
generation. The quality of electricity can be known as a set of values of parameters,
such as voltage magnitude, deviation in transient voltages and currents, continuity of
service, harmonics in the AC power and others. For IEEE Standard 1100-1999
Recommended Practice for Powering and Grounding Electronic Equipment, power
quality is the ideology of powering and grounding electronic equipment which is
acceptable with the functions of the equipment and compatible with the foundation
wiring system and other coupled equipment (IEEE, 2005). An ordinary classification
20
of power quality phenomena is also provided by IEEE based on the principal spectral
content, duration, and magnitude of the disturbance. Table 2.3 shows categories and
characteristics of power system electromagnetic.
Table 2.3: Categories and Characteristics of Power System Electromagnetic
Phenomena
No. Categories Typical Spectra
Content
Typical
Duration
Typical Voltage
Magnitude
1 Transients
1.1 Oscillatory
1.2.1 Low Frequency < 5 kHz 0.3 – 50ms 0 – 4 pu (per unit)
1.2.2 Medium Frequency 5-500 kHz 20µs 0 – 8 pu
2 Long Duration Variations
2.1 Interruption, Sustained >1 minute 0.0 pu
2.2 Undervoltages >1 minute 0.8 – 0.9 pu
2.3 Overvoltages >1 minute 1.1 – 1.2 pu
3 Voltage Unbalance Steady state 0.5 – 2%
4 Waveform Distortion
4.1 Harmonics 0 – 100th harmonic Steady state 0 – 20%
5 Voltage fluctuations < 25 Hz intermittent 0.1 - 7%
Source: IEEE Std 1159-1995
Blackouts, overheating transformers, circuit breaker trips, neutral conductors
overheating, and unexplained equipment downtime are the impacts of having poor
power quality in the electrical power system. The power quality occurs due to many
aspects. The most common technical issues can be caused by the integration of EV
and RE systems are voltage fluctuation, voltage rise, voltage sags, voltage unbalance,
reversed power flow, unnecessary neutral current flow, harmonics, and flickers.
Furthermore, voltage rise has been recognized as the most serious power quality
21
issues caused by the high penetration of renewable energy sources in UK (Lyons,
2010). Therefore, power quality issues are significant issues to be concerned.
2.3.1 Voltage Unbalance Factor (VUF)
In a three-phase balance power distribution system, the three line-neutral voltages are
equal in magnitude and are phase displaced from each other by 120 degree in a
balanced sinusoidal supply system. Any differences that exist in the three voltage
magnitudes and/or a shift in the phase separation from 120 degree is said to bring
about an unbalanced supply or familiarized as voltage unbalance (Gosbell, 2002).
The difference between a balanced system and unbalanced system is shown in Figure
2.8.
(a) (b)
Figure 2.8 (a): Phasor of the balanced three-phase voltage
(b): Phasor of the unbalanced three-phase voltage
Ideally, a three-phase power system should operate in balanced condition. However,
this condition is hardly to be attained as single-phase distributed generation is
increasing on the LV distribution networks. When the high penetration of small scale
22
embedded generation in the LV networks, voltage unbalances has become a serious
issue. This is due to the fact that the distributed small-scale embedded generation is
not centrally installed across the distribution networks. Moreover, the voltage
unbalance in the distribution networks can be caused by continuous varying of
instantaneous demand, irregular distribution of single-phase load across the three-
phase network, and unbalance or unstable utility supply.
There are two definitions for voltage unbalance. National Electrical Manufacturer
Association (NEMA) defines the voltage unbalance factor as the ratio of maximum
deviation of the three-phase line voltages to the mean of the three-phase
line voltages. According to the International Electrotechnical Commission (IEC)
standards, voltage unbalance factor (VUF) is defined as the ratio of negative
sequence voltage (V-) to positive sequence voltage (V+), as expressed below (P.
Trichakis, 2006; Lee, 1999):
The negative and positive sequence of the system voltage can be calculated by the
following expression:
c
b
a
V
V
V
aa
aa
V
V
V
2
2
2
1
0
1
1
111
3
where Va, Vb and Vc are the three-phase line voltages while V0, V+ and V- are the
zero, positive and negative sequence voltages components respectively. In Malaysia,
the statutory limit of the voltage unbalance is 2% (Bollen, 2002) which is much quite
23
similar as compared to 1.3% in the UK and 2% in EU (Trichakis, Taylor, Lyons, &
Hair, 2008).
It is important to maintain the VUF below the statutory limit. This is because when
voltage unbalance occurs, it will cause serious problem in distribution network where
there is a returned current flow through the neutral line which causes the increase of
power losses in the network and reduction in the efficiency of the distribution
network. Besides that, VUF also gives impact to the induction motor. According to
National Electrical Manufacturers Association (NEMA), the voltage unbalance of
1% at the terminal of fully loaded motor can create 6 -10 times the unbalance current.
Consequently, this unbalance current will cause the increase in heating by 6 to 10%
as shown in Figure 2.9. This can reduce the energy efficiency and lifespan of the
motors. As a result, the customers may need to replace the motor more frequently.
Besides generating unnecessary heat in the motor winding, voltage unbalance also
producing harmonic currents to the system, this affects the efficiency of the network.
Figure 2.9: Increase in motor heating due to voltage unbalance (Bishop, 2008)
24
Many researches have been undertaken regards the voltage unbalance mitigation. In
the paper of (Oeng & Sangwongwanich, 2016), strategy was proposed where
unbalanced voltage is mitigated by the injection of negative-sequence currents.
Simulation results show that 50% unbalanced voltage can be reduced using this
strategy. Besides, voltage unbalance caused by single phase PV inverters in LV
networks is reduced by 51% with the applied of DSTATCOM in (Shahnia, Ghosh,
Ledwich, & Zare, 2010). Voltage level is maintained at the point of connection by
injecting both the active and reactive power. Since single-phase PV systems worsen
the voltage unbalance ratio in the network, in (Bajo, Hashemi, Kjsær, Yang, &
Østergaard, 2015), three-phase PV inverters are controlled to inject and absorb the
real and reactive power in each of the phases. Results show that 45% reduction in
voltage unbalance can be attained. In (Chua, et al., 2012), single phase batteries were
used as ESS to mitigate the voltage unbalance and a reduction of 37% can be reached.
In the newer technology, electric vehicles are also used to mitigate the voltage
unbalance problem. Both active and reactive power are controlled in order to
compensate the uneven power flow in (Jiménez & García, 2012). The unbalance
voltage with a reduction of 27% can be reached with this method.
Since EV and PV systems are highly encouraged in Malaysia and majority of the PV
systems and EV are single-phase, voltage unbalance on the distribution network is
likely to happen in the future. It is believed that voltage unbalance would be a serious
problem when more and more EV and PV systems are integrated to the distribution
network.
25
2.3.2 Voltage Flickers
Voltage flickers are known as luminosity variation of lights which may influence the
human visual system depending on the frequency and the brightness. It is caused by
rapid voltage changes due to load changes in the power system. In overall, the
voltage flicker can be defined as an impression of unsteadiness of visual sensation
induced by a light stimulus, whose luminance or spectral distribution fluctuates with
time. Normally, the frequency is ranging from 1 Hz to 35 Hz when subject to change
in the illumination intensity. There is short term index (Pst) and long term flicker
index (Plt) for the measurement of the severity of the voltage flickers. For short term
index, it is computed by the fluctuating voltage over a period of 10 minutes, while
long term flicker index, also called the average of short term flicker, can be
calculated by the fluctuating voltage over a period of 2 hours.
Electrical furnace, welding machines, rolling mills, pumps and others are well known
as flicker generators. However, variable frequency drives (VFD), switching loads,
motor starting, and also variations in load current due to repetitive events such as
washing machines, refrigerators, dryers, air conditioners and others also causing
voltage flicker. On the other hand, the frequent switching of the PV inverters due to
the dynamic change of PV output also intensifies the issue of flickering.
The quality of voltage supplied to the networks is significantly reduced due to the
voltage flickers and these oscillations of voltage are kind of serious problems to the
networks. Voltage flicker may affect the starting torque, slip, starting current,
temperature rise, and even overloading of motors and generators. Furthermore,
26
domestic lighting, power electronic devices, computers and other devices also
affected by the voltage flickers, and reduce the life span of those fluorescent and
electronics devices.
2.3.3 Voltage Rises
Small-scale embedded generation (SSEG) such as photovoltaic modules and wind
turbines are becoming common on the low voltage distribution networks. However,
voltage rises are one of the issues that limit the penetration of SSEG. Simulation
studies using Power World simulation software demonstrate that the integration of
PV system is likely to increase the voltage level (Lewis, 2011). Voltage rises can
happen due to the reversed power flow with the severity depending on the system
impedance, generation location, generation characteristics and system load.
In Malaysia, the voltage tolerance boundary is set at a range of +10% and -6% for the
primary distribution networks with the voltage level of 240V. Any equipment will
perform satisfactorily if the voltage is always maintained within the statutory limits
by the network operators. There are different methods that can be used to mitigate
the voltage rise such as the generator active power curtailment, the wide area voltage
control, the reactive power management, and the energy storage system. The main
limitation for these approaches is the significant cost needed for the installation of
equipment, sensors, control systems, communication and other infrastructure changes.
In this project, the energy storage system is used to cater for the voltage rise issues to
27
ensure the voltage is within the voltage tolerance. It can store the excess power
during minimum load connected and discharge when it is peak load times.
2.3.4 Low Power Factor
Power factor is an index used to calculate the efficiency of electricity consumption,
which is a measure of how effectively the power is being converted into useful
output. The connection between the real and reactive power is expressed by the right
triangle shown in Figure 2.10 where real power is the horizontal axis and reactive
power is in vertical axis. The overall burden deposited on the grid by the load from
both its real and reactive parts is known as the apparent power.
Figure 2.10: Relationship between real, reactive and apparent power
Power factor is the cosine of the angle, PF = cos φ. The load is resistive when φ is
zero degrees and power factor is unity. If the load also includes reactive elements,
the reactive power will be non-zero and the apparent power will go beyond the real
power as shown in Figure 2.10. Hence, when two loads are given, each consuming
the same amount of real power, the one with the higher power factor will be draw
28
less circulating current and more efficient than the other with the lower power factor.
A poor power factor can be the result of inductive loads such as an induction motor,
lighting ballasts, power transformer, welder or induction furnace, or it can be due to
high harmonic content or distorted or discontinuous current waveform. This
composes a loss to both the consumer as well as the utility company. The losses in
the cable conductors at a power factor of 0.8 are 1.57 times the losses at unity power
factor, and the losses at a power factor of 0.4 will be 6.25 times that at unity power
factor (REEEP, 2006).
PV systems that supply power to the utility can actually deteriorate the power factor
after the system is connected to the grid (R. Romo, 2015). An example will be used
to show how PV system can affect the network power factor. In Figure 2.11, it shows
a network with a load connected to the power grid. The load draws 1000 kW of real
power and 450 kVar of reactive power which results in 1096 kVA of apparent power.
By using P = S cos φ or Q = S sin φ, φ can be calculated as 24.2º and hence, the
power factor is 0.912 as shown in figure 2.11. In Figure 2.12, a PV system is placed
to displace 50% of the real power from the power grid. Therefore, each of the power
grid and the PV system feeds 500 kW to the load. In fact, the PV inverter is designed
to operate with a unity power factor. The PV system with the unity power factor only
supplies real power to the grid. Hence, the power factor angle is increased to 42º
after the PV system is connected to the system. Therefore, the power factor drops to
0.743. This example shows clearly that the power factor drops after the PV system is
connected.
29
Figure 2.11: Normal power factor in the network
Figure 2.12: Power factor being reduced when the PV system is integrated
The inefficiencies associated with the low power factor means that the current along
the distribution lines are high. As a result, the utility company will impose power
factor penalty on the customers. The utility company must set minimum power factor
standards in accordance with applicable tariffs. In Malaysia, any customers with low
30
power factor will be penalised by Tenaga National Berhad (TNB). The power factor
penalty is calculated as follows:
(a) For every 0.01 less than 0.85 power factor, 1.5 % surcharge of the current bill
is imposed.
(b) For every 0.01 less than 0.75 power factor, 3 % surcharge of the current bill
is imposed.
For example, if the total unit used by a company is 5000 kWh per month at the rate
of RM 0.323/ kWh and the power factor is 0.5, the total electricity bill is calculated
as 5000 x RM 0.323 or RM 1615. The monthly surcharge on the low power factor is
(0.85 – 0.75) x 1.5 + (0.75 – 0.5) x 3 = 0.9 due to the rate stated above. Hence, low
power factor penalty is calculated as 0.9 x RM 1615 or RM 1453.5. Therefore, the
total amount of the electricity bill for the month is RM 1615 + RM 1453.5 = RM
3068.5. In this example, the low power factor causes 90% increase of its original bill.
In reality, PV power output is not constant and fluctuating throughout the day
depending on the clouds and position of the sun. Fluctuating power factor will have a
high risk to get the power factor penalty as the low power factor surcharge is charged
based on the amount of monthly reactive energy consumption. This is because the
average power factor over the month is calculated by 22 QP
P
, where P is the
active energy consumption and Q is the reactive energy consumption. If the average
power factor drops below the standards, power factor needs to be increased to
improve the network efficiency. The supply of the reactive power from the grid has
to be reduced in order to attain the unity power factor. Therefore, customers are
advised to install KVAR generators such as, capacitor, corrector, synchronous
31
generators, and synchronous motors. There are a number of solutions available to
maintain the network reactive power such as static VAR compensator, static
synchronous compensator and active filters (Li, Zhang, 2008; Wu & Liu, 2011;
Muhammad A. Saqib, 2015).
2.4 Solutions for Mitigating the Power Quality Issues
The increasing amount of renewable energy penetration to the electrical networks
will raise the power quality issues. Besides, electric vehicles are also one of the
significant elements to affect the power quality of the conventional electrical systems.
Voltage rise, voltage unbalance, voltage sags, voltage regulation, power factor are
the common technical issues caused by the integration of renewable energy and
electric vehicles. Demand side management, renewable energy curtailment, Flexible
AC Transmission System (FACTS), energy storage, and on-load-tap-changer control
are the existing strategies used to improve the power quality on the distribution
networks.
2.4.1 Demand Side Management
Demand side management (DSM) refers to actions taken on the consumer’s sides to
change the amount of the electricity usage. One of the main functions of DSM is to
manage the load in order to change the power demand profile of customers
(CHUANG & GELLINGS, 2008). DSM programs consists of planning, scheduling,
32
monitoring and executing activities of power demand for the objectives of
minimizing the electricity prices, reducing the energy infrastructure investments,
enhancing the energy security and others ancillary advantages. There are four basic
schemes of DSM as follow (Johnson, 2007):
1. Peak Demand Clipping: Decrease the peak demand
2. Energy Efficiency and Conservation: reduce overall energy usage
3. Load Building: Maximize demand to off-peak hours
4. Load Shifting: Relocate demand to off-peak hours
The DSM strategies have the objectives of increasing end-use effectiveness to
prevent or delay the construction of new generation plants (Lim Yun Seng, 2006).
Besides, the DSM is also a series of interconnected and flexible programs whereby
customers play a great role to decide their own demand for electricity during peak
time and cut down their energy consumption in overall (Dabur, Singh, & Yadav,
2012). In (Tande, 2000), the DSM can be used as an effective method to mitigate the
power quality issues due to the wind energy integration. The load controller is used
to control the loads in order to match with the wind power generation to mitigate the
voltage quality issues with the minimum constraints on the renewable energy sources.
Besides, the authors of (Lim, WHITE, NICHOLSON, & TAYLOR, 2005) states that
DSM provides a robust solution to mitigate voltage rise issues on a LV distribution
network with a large amount of renewable energy sources. However, DSM is not
popular in Malaysia. There is insufficient of controlled loads for demand side
management to function effectively. In addition, the DSM is highly relies on costly
and proprietary technology solutions. Hence, the impact of various DSM programs
33
should be investigated. Besides, the downside of the DSM is that the customers’
acceptance is still very low. Therefore, it is difficult to implement DSM efficiency.
In the paper of (Ning, Luis, & Daniel, 2011), EV is ideal appliances for DSM scheme.
This is because the interruption of the charging of an electric vehicle for a few
minutes is not a major inconvenience. Hence, electric vehicles will be the first loads
to be inhibited during peak time to reduce the peak load and hence, reduce the
voltage rise of the networks.
2.4.2 Static Synchronous Compensator (STATCOM)
STATCOM is one of the flexible AC transmission system (FACTS) devices. It is a
shunt compensator to control the real and reactive power flows by injecting the
current with a controllable angle. It is widely used to control the power factor,
regulate the voltage, stabilize the power flow, and improve the performance of the
power systems. Under balanced condition, STATCOM works well, whereas a large
amount of negative sequence current will be induced when it goes through the
STATCOM under unbalanced condition. Therefore, in the paper of (Li, Liu, Wang,
& Wei, 2007), novel strategies are proposed to compensate the utility voltage
unbalance using the STATCOM. In (Hendri Masdi, 2004), the designed D-
STATCOM which is the distribution static compensator can mitigate the voltage sags
caused by the three-phase balanced fault. It is the matter of the amount of reactive
current needed to be compensated by the D-STATCOM.
34
The traditional STATCOM are limited to improve the system stability as it has no
significant active power capability. Nevertheless, in the paper of (Aarathi A.R, 2014),
the usage of STATCOM integrated with super capacitor energy storage system able
to enhance the system stability with the grid connected photovoltaic system. In
another similar paper, (Xi, Parkhideh, & Bhattacharya, 2008) also states that the D-
STATCOM integrated with ultracapacitor, which is kind of energy storage, is suited
to mitigate the voltage regulation and voltage sags in the distribution systems. In (E.
Ghahremani, 2014), the authors also state that the combination of ESS and
STATCOM would add the benefits as individual such as reduced power flows,
increased system loadability, lower the transmission line losses, and improve the
power system stability. However, the STATCOM is always an expensive option for
reactive power control (T. Aziz, 2013).
2.4.3 Energy Storage System
At the moment, there are different mechanisms available for the storage technologies.
They can be classified into two categories: short term discharge energy storage
devices and long term discharge energy storage devices. For the short term discharge,
it responses fast to the power system needs. These kinds of energy storage devices
are normally applied in improving the power quality issues, to cover the load during
start up, synchronization of the backup generators and to compensate the dynamic
response of renewable power sources (Tande, 2003). For long term discharge, it is
able to supply power from few seconds to few hours. Their response is usually
slower than the short term discharge. It is usually applied on energy management,
35
renewable energy sources integration and power grid congestion management (Price,
Bartley, Male, & Cooley, 1999). Batteries that used commonly are considered as
long term discharge devices.
The energy storage system is an expensive option as batteries in this stage of
development is expensive, especially those using lithium ion batteries. Currently,
lead acid batteries are cheaper than lithium ion batteries. However, the lead acid
batteries have relatively short lifespan and its life is reduced when depth of charge
and discharge are too high. Nevertheless, it can avoid the renewable power
curtailment and reversed power flow (Zito, 2010). In addition, research has been
carried out to determine the feasibility of the energy storage system to improve the
power quality and provide ancillary services to the electrical networks (Nguyen and
Flueck, 2012).
Energy storage systems can be used for many other applications, such as load
levelling, load shifting, peak clipping and others. On the other hand, the energy
storage system can offer additional benefits in utility settings because it can decouple
demand from supply, hence mitigating the frequency deviations and allowing
increased asset utilization, facilitating the penetration of renewable sources, and
improving the reliability, flexibility, and efficiency of the electrical network
(Schoenung, 1996).
The ability to store the energy for later use is important in many circumstances such
as the variable and uncontrollable power generation of renewable energy. In the
paper (Andreas Poulikkas, 2014), the storage system is introduced to mitigate the
36
problem of overvoltage due to the PV integration to the LV distribution feeder of
Cyprus. Apart from that, the energy storage can provide power whenever the grid
reaches its peak demand due to large amounts of plugged-in EV or due to the low
penetration of RE to reduce the voltage sags occurred in the networks. The authors of
(T. Shigematsu, 2002) states that voltage sags can be mitigated and fluctuating wind
power output can be stabilized by using the energy storage system. Hence, the energy
storage has smoothing function when the power is not constant. Besides, voltage rise
and voltage unbalance caused by intermittent PV can be mitigated using novel fuzzy
controlled energy storage on LV distribution networks stated in (Wong, Lim, &
Morris, 2014).
An appropriate control strategy of the ESS is important in mitigating the power
quality issues. Without any control algorithm used, ESS was controlled manually to
obtain the optimum result in the paper of (Chua, et al., 2012). VUF was reduced from
1.6% to 0.8%. It is a 50% reduction in VUF. However, in the paper of (Chua, et al.,
2012), the VUF mitigation in low voltage distribution networks with photovoltaic is
done with the energy storage system using the PI controller. The results showed that
the VUF was reduced from 1.4% to 0.9%, which is a 35.7% drop in VUF. With the
fuzzy control strategy in (Wong, Lim, & Morris, 2014), VUF was reduced from
1.9% to 0.9%, which is a VUF mitigation of 52.6% in 5 minutes. Hence, different
control strategy of ESS can vary the efficiency in the mitigation of VUF. In this
project, fuzzy control algorithm is used as it has the higher percentage in the VUF
mitigation.
37
In conclusion, the integration of energy storage appears as a very positive solution
for the power unbalance mitigation. It can act as a frequency control at the
transmission level, voltage control and capacity support at the distribution level.
Besides, it can be installed to provide peak shaving at the consumer level. Also, the
energy storage system can be used to mitigate the power quality issues caused by
renewable energy and electric vehicles provided it is equipped with an appropriate
control strategy.
2.5 Summary
The government has launched variety of incentives and programmes to promote the
penetration of renewable energy and electric vehicles to reduce the greenhouse gas
emissions. However, power quality issues such as voltage rise, voltage unbalance,
power factor and voltage sags influence the performance of the electrical network
due to the large integration of distributed renewable energy sources and plugged-in
electric vehicles. The existing methods on mitigating power quality issues might not
be economically attractive because the super-capacitor, STATCOM and SVC are too
expensive. In addition, they are generally used in transmission system rather than LV
distribution networks to cater the problems. The active management is also one of
the solutions to overcome the technical issues as it can respond based on the network
conditions. Since the energy storage system is also considered as an active
management system, the energy storage system can be an appropriate solution for
mitigating these power quality issues due to the penetration of renewable energies
and electric vehicles.
38
CHAPTER 3
ARCHITECTURE OF THE EXPERIMENTAL LV NETWORK
3.1 Introduction
The Malaysian LV distribution network is a radial three-phase four-wire system with
voltages of 33 kV, 11 kV, and 415 V. A laboratory scale experimental LV
distribution network is designed to emulate the real LV distribution network in
Malaysia. This is to investigate the effects of the penetration of RE and EV on the
distribution networks. In this chapter, the design and each of the elements of the
experimental LV distribution network are explained. The physical architecture of the
proposed energy storage system is also presented in this chapter
3.2 Experimental LV Distribution Network
An experimental LV distribution network is developed to investigate the
effectiveness of the fuzzy controller as shown in Figure 3.1. This network is a Terra-
Terra (T-T) earthing three-phase four-wire system. It consists of a network emulator,
two single-phase 3.6 kWp PV systems, a 2.0 kW single-phase wind turbine emulator,
a 9 kW three-phase load emulator, an electric vehicle and the fuzzy controlled energy
39
storage system. Each of the devices will be further elaborated in the following
subsection.
Bb
Figure 3.1: Experimental network diagram
Figure 3.2 shows the flexible LV panel. All the loads, electric vehicle, energy storage
system, wind turbine emulator and PV are connected to the LV network through this
LV panel via 5-pin plug. There are phase selector switches installed for PV systems,
wind emulator systems, and electric vehicles. Therefore, these single-phase devices
can be switched to other phases according to the condition. All the power meters are
mounted on this LV panel for monitoring purposes. The main circuit breaker is used
to protect the load bank from overcurrent and short circuit. Moreover, overcurrent
protection and residual current circuit breaker are installed in the panel for safety and
protection purpose and to protect the equipment. There is also an emergency stop to
disconnect the incoming power from the grid to avoid any hazard accident to happen.
Figure 3.3 shows the experimental setup in the laboratory.
40
Figure 3.2: Flexible LV panel
Figure 3.3: Experimental setup
41
3.2.1 Network Emulator
A network emulator is used to represent the LV distribution network in Malaysia.
The network emulator is a 15 kVA synchronous generator coupled to an induction
machine controlled by a variable speed drive (VSD) as shown in Figure 3.4. Figure
3.5 shows the VSD of ABB with model number of ACS800-01-0025-3+E200+K458.
Figure 3.6 is the synchronous generator coupled to an induction motor. The
specification of the VSD is also shown in Table 3.1. The system frequency can be
varied by changing the reference speed of the VSD. The VSD drives the induction
motor at 1500 rpm. Hence, the system frequency is regulated at 50 Hz, 240 V for the
experiments. The network emulator can provide power balance over the three-phase
system. With this design, it can decouple the experimental network emulator from
the utility grid. This network emulator is a three-phase four wire configuration with
T-T earthing system which is a common configuration in Malaysia.
Figure 3.4: Grid emulator
From
Grid
Variable
Speed Drive
Three Phase
Induction
Motor
Three-Phase
Synchronous
Motor
To Experimental
Network
42
Figure 3.5: Variable speed drive used to drive the induction motor
Figure 3.6: Synchronous generator coupled to an induction motor
Table 3.1: Specifications of VSD for network emulator
Characteristics Specifications
Input Voltage 3~ 380…415 V
Input Current 42 A
Input Frequency 48…63 Hz
Output Voltage 3~ 0…UInput V
Output Current 44 A
Output Frequency 0…300 Hz
43
3.2.2 Photovoltaic Systems
Solar panels that used in this experiment are built by 32 panels of polycrystalline PV
modules, a product of Panasonic with model number VBMS230AE01. These panels
are categorized to 4 strings, 8 panels in a string. From the specifications of the PV
modules in Table 3.2, each module can contribute a maximum power of 230 W.
Therefore, each string will have the capacity of 1.84 kWp and the total capacity of
the PV system is 7.36 kWp. The PV modules are oriented 5º tilt to face north. This
allow the dust on the PV modules be washed by the rain water and allow sunlight to
penetrate on the PV modules throughout the day. Figure 3.7 shows the line diagram
which PV modules are connected to the 3.88 kWp PV inverter named Sunny Boy
Transformerless Inverter, SB 3600TL-21. Figure 3.8 is the solar panel mounted on
the 3.3 m height metal structure in the real experimental setup.
Table 3.2: Specification of PV modules
No Characteristics Specifications
1 Maximum power (Pmax) 230 W
2 Short circuit current (Isc) 8.42 A
3 Open circuit voltage (Vac) 37 V
4 Maximum power current (Ipmax) 7.83 A
5 Maximum power voltage (Vpmax) 29.4 V
6 Maximum system open circuit voltage 1000 V
44
Figure 3.7: PV system
Figure 3.8: Photovoltaic panel mounted on a metal structure
3.2.3 Wind Turbine Emulator
The wind turbine emulator is consisted of a three-phase induction motor coupled
with a three-phase synchronous machine, a variable speed drive, a rectifier and a
single-phase inverter known as SMA Windy Boy 2500. This wind turbine emulation
system is illustrated in Figure 3.9. The variable speed drive (VSD) is used to control
the speed and torque of the induction machine in order to emulate the actual speed
and torque characteristics of the wind turbine. The VSD from ABB with model
45
number of ACS800-01-0005-3+E200+K458 is used in this project as shown in
Figure 3.10 (b). Figure 3.10 (a) is the induction motor coupled with the synchronous
machine which is used as the emulator. The specification of the VSD is also shown
in Table 3.3. The three-phase output is then rectified using a rectifier to supply DC
voltage to the Windy Boy. Due to the specification of Windy Boy in Table 3.4, it will
provide 50 Hz, single phase AC voltage with nominal voltage of 230 Vac to one of
the phases of the experimental LV distribution network. Figure 3.11 (a) and (b) show
the rectifier and SMA Windy Boy used in the wind turbine emulation system.
Figure 3.9: Topology of the wind turbine emulation system
(a) (b)
Figure 3.10: (a) Induction motor coupled with synchronous machine driven by (b)
VSD of wind turbine emulator
46
Table 3.3: Specifications of VSD for wind emulator
Characteristics Specifications
Input Voltage 3~ 380…415 V
Input Current 7.9 A
Input Frequency 48…63 Hz
Output Voltage 3~ 0…UInput V
Output Current 8.5 A
Output Frequency 0…300 Hz
Table 3.4: Specifications of SMA Windy Boy 2500
Characteristics Specifications
VDC max 600 V VDC MPP 224 – 480 V
IDC max 12 A
VAC nom 230 V
fAC nom 50/60 Hz
PAC nom 2300 W
IAC nom 10 A
cos φ 1
(a) (b)
Figure 3.11: (a) Rectifier; (b) SMA Windy Boy 2500
47
3.2.4 Electric Vehicles
The electric vehicle used is the Proton Wira converted by UTAR students as shown
in Figure 3.12. This electric vehicle is designed to use three-phase AC motor and the
speed of motor is controlled using the motor controller of Curtis 1239-8501. The
motor can be operated at high voltage ranging from 144 V to 170 V. When the
current is increased to 500 A, the speed of the motor can reach up to 88 horsepower
and 108 lbs of torque. The maximum current output rating of the battery charger used
is 15 A, having 93 % of efficiency. When the electric vehicle is plugged to the
network, it can be charged up to 12 A practically. It is connected to the network
through the LV panel via the 5-pin plug. Figure 3.13 shows the 5-pin connector of
the electric vehicle. In this experiment, electric vehicle acts as a significant load to
absorb power from the network and causing voltage to drop significantly.
Figure 3.12: Electric vehicle used in the experiment
48
Figure 3.13: Charging connector of the electric vehicle
3.2.5 Load Emulator
Three phase load bank in Figure 3.14 is designed and built to emulate the consumer
load. The load bank consists of 18 units of 500 W power resistors. Each resistor can
be controlled by triggering the solid state relay (SSR) and the relay is controlled via
Labview control system using NI 9403 Digital Output Module as interface between
the supervisory computer (PC) and the controllable load. This NI 9403 is a 37
channel, 7 µs bidirectional digital I/O module for NI Compact DAQ chassis. Each
channel is compatible with 5 V/TTL signals and providing 32 digital input or output
channels as shown in Figure 3.15. The COM lines will be referred as the common
point for the signals. The 5 V digital signals will be channelled to the D2425 SSR as
shown in Figure 3.16. This D2425 SSR is silicon controlled rectifier output which is
suitable for heavy industrial loads and it is operated with zero switching. The control
voltage is 3-32 VDC and the current rating is 25 A. Figure 3.17 shows the
configuration of the controllable load bank. Each phase can be loaded up to 3000 W
49
with an increment of 500 W per step. However, there is limitation for the designed
load bank as it is purely resistive. Hence, the load bank is having the unity power
factor all the way. In order to emulate a better consumer load, inductive load is
connected.
Figure 3.14: Load bank used in the experimental
Figure 3.15: NI 9403 37 channels connection diagram
50
Figure 3.16: D2425 Solid State Relay
Figure 3.17: Arrangement of controllable load bank
Inductive load used in the experiment is the laboratory equipment from
ELWE Germany as shown in Figure 3.18. It is a three-phase variable inductive load
consisted of 8 steps. It has 45 - 360 Var and the current rating is 0.2 - 1.56 A.
Applying the rating given, inductance calculated using the formula P=IV, VL=ILXL
and XL=2πfL, inductance, L, is in the range of 0.47 - 3.58 H. However, based on the
measurement, inductance calculated is in the range of 0.339 - 2 H. Hence, this
inductor load can only contribute maximum of 259 Var for each phase during the
51
experiment. This inductive load is used to vary the power factor of the emulated
network. Power factor will be 0.85 when inductive load is switched to 7 steps.
Figure 3.18: Inductive load used in the experiment
3.2.6 Energy Storage System
The proposed energy storage system consists of three bi-directional inverters. Each
bi-directional inverter is integrated with four battery banks. A RS232/RS485
converter is the communication link between the supervisory computer and the bi-
directional inverter. The structure of the proposed energy storage system is shown in
Figure 3.19. The Sunny Island is able to be controlled to maintain the voltage and
frequency of the grid at a constant level to mitigate the voltage rise, voltage
unbalance and improve the power factor of the networks. This energy storage system
52
is usually installed at the loads because RE and EV are usually placed at the
customer sides.
Figure 3.19: Setup of the proposed energy storage system
3.2.6.1 SMA Sunny Island 5048
Sunny Island 5048 is a bi-directional inverter produced by SMA GmbH, Germany.
The most advanced technology used in their Sunny Boys is combined with an
outstanding feature set for off-grid PV applications to form the SMA’s next
generation of Sunny Island inverters. The new Sunny Island 5048 shown in Figure
3.20 has incredible surge capabilities and excellent AC output characteristics. It
provides a very clean and stable 230 VAC output. Multiple units can be used
together for single-phase or three-phase applications. In this project, three Sunny
Islands are used for each phases. The power provided by the Sunny Island can be
increased by simply adding an additional Sunny Island inverter in parallel. The
Sunny Island also has the special ability to facilitate AC coupling of all the power
53
generation sources. This is kind of efficient use of available AC power and greatly
simplifies the system expansion. The overall life of the batteries is optimized with
the help of the Sunny Island 5048 through its advanced battery management system.
Estimation of state of charge of the batteries can be obtained by this battery
management system as criteria for the energy storage system to charge and discharge.
Figure 3.20: Sunny Island 5048 bi-directional inverter
One of the greatest advantages of AC coupling is that it allows any type of power
generating device to be integrated to the system, such as solar, wind, hydro, even gas
or diesel powered generators. Hence, the chance to take advantage of all the
resources available at different location is offered to the system designers. It also
allows for easy system expansion by providing a solution for connecting future
power sources. AC coupling offers simple system expansion and guarantees that the
system will continue to serve the needs of its users safely and reliably over the long
term.
54
As a conclusion for Sunny Island 5048, the settings needed for the operation can be
configured in few steps. It is flexible in its application, extendable and takes on all
the control processes. In addition, the device is highly efficient and has an ergonomic
die-cast aluminum enclosure. For systems, it is flexible from 3 kW to 300 kW, single
and three-phase operation. It is connectable in parallel and modularly extendable, AC
and DC coupling. It is simple and easy commissioning with the “Quick
Configuration Guide” and complete off-grid management. It has high efficiency,
intelligent battery management for maximum battery life, charge level calculation to
obtain the condition of battery. It has extreme overload capacity, OptiCool active
cooling system and 2-year SMA warranty. Sunny Island also offers the highest
flexibility where real time active and reactive power controls are possible. With this
features, the Sunny Island 5048 integrated with batteries are controlled by the
proposed fuzzy controller to mitigate the voltage rise, voltage unbalance and improve
the power factor caused by the intermittent RE power output on LV distribution
networks. Table 3.5 shows the specifications of SMA Sunny Island 5048 used in this
project.
Table 3.5: Specifications of SMA Sunny Island 5048
Output data
Nominal AC voltage (adjustable) 230 V (202 V – 253 V)
Grid frequency adjustable 45 Hz – 55 Hz
Continuous AC output at 25 °C / 45 °C 5000 W / 4000 W
Continuous AC output at 25 °C for 30 / 5
/ 1 min 6500 W / 7200 W / 8400 W
Nominal AC current 21 A
Max. current 100 A (for 100 ms)
Output voltage harmonic distortion factor < 3 %
Power factor –1 to +1
Input data
Input voltage (range) 230 V (172.5 V – 250 V)
Input frequency 50 Hz (40 Hz – 70 Hz)
55
Max. AC input current (adjustable) 56 A (2 A – 56 A)
Max. input power 6.7 kW /12.8 kW
Battery data
Battery voltage (range) 48 V
Max. battery charging current 120 A
Continuous charging current 100 A
Battery capacity 100 Ah to 10000 Ah
Charge control IUoU with automatic full and equalization
charge
3.2.6.2 Batteries
Batteries store energy in electrochemical form, creating electrically charged ions.
However, batteries consist of varieties of technologies, applying different operation
principles and materials. The most important battery concepts are electrochemical
and redox flow. For electrochemical batteries, there are lead acid, nickel cadmium,
nickel metal-hydride, sodium sulphur, lithium ion and others. Lead acid batteries are
commonly used due to its reasonable price.
Battery used in this project is Hoppecke solar.bloc valve-regulated lead acid (VRLA)
battery for typical applications of solar or off grid applications, storage for direct
consumption of photovoltaic energy, telecommunications and traffic systems. This
VRLA battery is reported to have a higher operational efficiency and longer lifetime.
Figure 3.21 shows the batteries used in the experiment. Each battery bank consists of
four VRLA batteries rated at 48 V, 115 Ah. Four banks are connected in parallel in
order to increase the capacity to 460 Ah. The nominal voltage of the battery is 12 V
which is having a total capacity of 5.52 kWh. The rated power for the Sunny Island
used is 4.0 kW at 45°C and 5.0 kW at 25°C. Figure 3.22 shows the connection of the
56
energy storage system. This energy storage system acts as one of the energy sources
in the experiment.
In terms of sizing of the battery, assumed that 400 W demand is introduced in the
network. Using the equation 3. 1, the annual energy consumption is 3504 kWh, while
the efficiency is assumed to be 90%, battery size needed for a 1-day supply is 222.22
Ah. In the experiment, 460 Ah battery is connected, however, only 230 Ah is useable
due to its 50% of depth of discharge. Hence, the battery used in the experiment is
able to sustain the 400 W load in a day, assuming there is no any PV power output
for the day. In order to sustain for a week, the battery capacity needed is 1555.55 Ah.
Hence, the rated battery capacity has to be above 3111.1 Ah in order to sustain for a
week.
----- (3.1)
57
Figure 3.21: Batteries
Figure 3.22: Arrangement of energy storage system
The main benefit of the lead–acid battery is its low cost; the main drawbacks are its
large size and weight for a given capacity and voltage (Buchmann, 2009). Lead–acid
batteries should never be discharged to below 20% of their full capacity, because
internal resistance will cause heat and damage when they are recharged. Therefore,
50% is the recommended state of charge (SOC) for the batteries. Figure 3.23 shows
the estimated life cycles based on depth of discharge for C10 Hoppecke solar.bloc,
which is the batteries used in the experiment (Hoppecke, 2012). It is observed that,
1,193 life cycles can be obtained with 48% depth of discharge.
58
Figure 3.23: Depth of discharge for C10 Hoppecke solar.bloc
The higher the conversion efficiency, the less energy is converted into heat and
hence, the faster a battery can be charged without overheating. The conversion
efficiency of the batteries is better when the batteries are having lower internal
resistance. One of the main reasons why lead-acid batteries monopolize the energy
storage markets is that the conversion efficiency of lead-acid cells at 85%-95%
which is much higher than Nickel-Cadmium (NiCad) at 65%, Alkaline (NiFe) at
60%, or other cheaper battery technologies. Lead-acid is moderately expensive,
having moderate energy density, and moderate rate of self-discharge. Higher
discharge rates result in considerable loss of capacity. So, lead acid batteries are used
in the project for charge and discharge purpose.
59
3.2.7 Communication and Data Acquisition Using LabVIEW
A data acquisition system is developed for monitoring the power flow of the network.
Several 40/5 A current transformers are used to measure current while NI 9225 3
Channels Voltage Measurement Card is used for voltage measurement. Elcontrol
STAR 3 power meters are used to measure and display the parameters. All the three-
phase voltage and current at the point of connection around the LV distribution
network are monitored. The data acquisition device used is the CompactDAQ 9174
Data Acquisition Chassis together with the NI 9225, a 3 Channels Voltage
Measurement Card module. Figure 3.24 shows the CompactDAQ 9174. The chassis
is connected to the PC to monitor the RMS voltage via Laboratory Virtual Instrument
Engineering Workbench (LabVIEW).
Figure 3.24: CompactDAQ 9174 Data Acquisition Chassis
Another data acquisition device used is the Data Taker DT82 as shown in Figure
3.25. It is a stand-alone and real time data logger which is used to acquire every data
entry. It can communicate with PC through Ethernet, USB, or RS232. Ethernet is
used in this project. A web based program developed by the manufacturer can be
accessed by the internet protocol (IP) address assigned to the data logger. Therefore,
Data Taker is configured to acquire the data needed and the PC is used to record the
60
data collected via LabVIEW. The program written in the data logger to acquire data
from the power meters is shown in Appendix A.
Figure 3.25: Data Taker DT82 Intelligent Data Logger
The PC is connected to the three single-phase energy storage system using RS232/
RS485 converter of ICP CON 7563. Signals sent are interpreted into a standard open
process control (OPC) format via YAOPC server which is under SMA possession
protocol. Hence, the energy storage system can be controlled through the control
algorithm developed in LabVIEW. Figure 3.26 shows the flow of data measurement
and the instruction between PC and the three-phase energy storage system.
Figure 3.26: Data acquisition between PC, Data Taker, cDAQ 9174, and the bi-
directional inverter
61
LabVIEW is used in this project to develop the control algorithm. The programming
language applied in LabVIEW is a dataflow programming language. By connecting
the wires, the structure of a graphical block diagram on different function-nodes is
used to control the execution. Variables are propagated with these wires and all the
nodes can be carried out once all the input information is ready. It is capable to
execute in parallel since multiple nodes can be run at the same time.
The configuration of user interfaces is tied into the development cycle which known
as a front panel. The LabVIEW programs are named as virtual instruments (VIs).
There are three elements in each VI, namely: front panel, block diagram, and
connector panel. The controls and indicators on the front panel can input data or
export data from an executing VI. The VI can be used as a subVI. Hence, a VI can
either be executed as a normal program with its front panel or dropped as a node onto
the block diagram as subVI. The connector panel is used to determine the inputs and
outputs node. This means that each VI can be easily tested and troubleshooting work
can be done before being embedded into a larger program.
62
Figure 3.27: Screenshot of LabVIEW program
Figure 3.27 is a LabVIEW program illustrating the necessary elements in LabVIEW
programming, which is the dataflow of block diagram in the right frame and the I/O
variables in graphical objects in the left frame. The entire algorithm will be designed
and constructed in the block diagram while the front panel is the graphical user
interface presented.
The main benefit of LabVIEW as compared to other programming tools such as C++,
C#, and FORTRAN, is their wide-ranging support for accessing instrumentation
hardware. It also contains drivers and abstraction layers for many different types of
instruments and buses. The abstraction layers provide standard software interfaces to
communicate with the hardware devices. Hence, the provided driver interfaces can
save the program development time. The National Instrument is much user friendly
as compared to other traditional or competing systems because people who are
having limited knowledge in writing a program can write programs and set up test
solutions in a shorter time period with LabVIEW.
63
3.3 Summary
The LV distribution network is set up to study the loading and generation conditions.
This network consists of a network emulator, photovoltaic systems, wind turbine
emulation systems, electric vehicles, load emulator, and energy storage system with
SMA Sunny Island 5048 and batteries. The network emulator is a 15 KVA
synchronous machine coupled with an induction motor driven by a variable speed
drive to regulate the system at 240 V, 50 Hz. Two PV systems, a wind emulator and
an electric vehicle are installed on the experimental LV network. All the devices are
single-phase and phases to connect can be selected freely by switching the phase
selector. Besides, a 9.0 kW three-phase controllable purely resistive load bank is
developed. A laboratory equipment of three-phase inductive load is also used to vary
the power factor of the networks. The data acquisition system is formed by several
current transformers, power meters, data logger, NI CompactDAQ chassis with
measurement cards, and a supervisory computer. All the measurements are displayed
and recorded using LabVIEW. Performance of the experimental LV distribution
network is tested. It is similar to the Malaysia’s LV distribution networks.
64
CHAPTER 4
Design of Control Algorithm
4.1 Introduction
The details of the control strategy of the proposed fuzzy controlled energy storage
system are presented in this chapter. Fuzzy logic is selected as the strategy of the
controller because it can produce an appropriate output based on the needs and
changes of the network parameters, which is different from the conventional Boolean
method as controller. Hence, the amount of real and reactive power can be
manipulated based on the changes of network parameters effectively. The coding in
LabVIEW is also explained in this chapter.
4.2 Fuzzy Controller for the Energy Storage System
Load demand, resistance, and reactance of the distribution network are usually
unknown and varying time by time, imply voltage unbalance, voltage fluctuation and
power factor also varying simultaneously. Particular measure of real power that
energy storage system needed to supply or absorb, to maintain the network voltage
65
level, is hard to be estimated by using the analytical method as voltage magnitude is
dynamic. The equation of voltage variation is expressed as follow:
---------- (4.1)
where R and X are the line resistance and reactance, Vnominal is the rated voltage at
the PCC. While PPV-L and QPV-L are denoted as
---------- (4.2)
---------- (4.3)
where PLoad and QLoad are the real and reactive power of the load, PPV and QPV are the
real and reactive power output of PV, PInverter and QInverter are the real and reactive
power output of the bi-directional inverter. Voltage at the PCC is maintained within
the tolerance boundary when PInverter and QInverter are determined provided other
parameters are recognized. However, it is unknown for R and X. In addition, PInverter
and QInverter is in constantly changing power flow condition. Therefore, fuzzy logic
control is much appropriate to solve the dynamic voltage problems due to uncertainty
of parameters needed.
The extended fuzzy control algorithm is developed using LabVIEW in a PC which is
linked to the bi-directional inverters. Real time measurements will be collected and
manipulated by the controller such that an appropriate instruction can be sent to the
inverters for any remedial actions. The real and reactive power output of the energy
storage system can influence the network voltage. However, real power is more
influential than the reactive power on the distribution network because the resistance
is greater than the reactance of the lines. Therefore, real power is used to correct the
66
voltage on the network. Below is the equation showing how the voltage change (∆V)
is related to the real (P) and reactive power (Q) output of the energy storage (Jenkins,
2000).
V
RP
V
XQRPV
---------- (4.5)
where R and X are the resistance and reactance of the distribution line. On the other
hand, reactive power is used to correct the power factor because the resistance is
greater than the reactance of the distribution lines as shown in Eq. (4.6) (Jenkins,
2000). However, the real power can still influence the power factor because the
amount of real power involved to correct the voltage is significant. The power factor
correction cannot be achieved satisfactory if it is carried out simultaneously with the
voltage correction. This is because the energy storage is not able to identify the right
amount of reactive power to be injected to correct the power factor if the energy
storage is still adjusting its real power injection. Therefore, it is proposed to correct
the voltage before the power factor.
V
RQXPV
---------- (4.6)
Figure 4.1 shows the flow chart of the control algorithm. The controller begins with
the collection of the measured three individual phase voltages (Vphx) and three
individual phase power factors (PFphx) in real time from the monitoring system. Vphx
is the measured voltage at the PCC. The controller will check whether the voltage is
within the acceptable range of 238 V and 243 V or not. Park's transformation is used
to determine the phase with the maximum voltage deviation. Voltage deviation, ΔVy,
is computed by using Vyphx–Vref, where Vyphx is the measured voltage of the most
severe phase at the PCC and Vref is the nominal voltage rated at 240 V.
67
The fuzzy controller uses ΔVy as one of the linguistic variables to create seven
linguistic terms through fuzzification in order to generate one output (ΔIp) as the
instruction to the bi-directional inverter through defuzzification. The bi-directional
inverter will adjust its real power output accordingly. This process will continue until
the voltage magnitude is within the statutory limit. Once the voltage magnitude is
restored, the voltage unbalance factor will also reduced.
Figure 4.1: Flowchart of control algorithm
Figure 4.1 shows that the controller will correct the power factor if the voltage
magnitude is with the acceptable range of 238V and 243 V. If the voltage is out of
the acceptable range of 238 and 243 V, then the controller will correct the voltage
first before the power factor. Similarly, should the measured power factor is below
68
0.98, ΔPFy is computed by using |PFyphx-1|, where PFyphx is the power factor of the
severe phase at the PCC. The fuzzy controller uses ΔPFy in the fuzzy system to
create one output (ΔIq) as the instruction to the bi-directional inverter. The instruction
Iq is used to manipulate the reactive power flow of the energy storage system in order
to improve the power factor of the network. The details of how ΔIp and ΔIq are
generated will be discussed in the following. The meaning of the inequality of
PFyphx(t) is to check whether the power factor is leading or lagging at time t. If
PFyphx(t) is greater than 0, then the power factor is leading. If it is not greater than 0,
then the power factor is lagging.
4.2.1 Details of the Fuzzy Controller
As mentioned above, ΔVy used as the input parameter for the predefined fuzzy
membership includes seven linguistic terms of voltage condition. It is necessary to
convert the voltage deviation and power factor into linguistic variables. This is
because the fuzzy controller takes the linguistic variables as the inputs that describe
the conditions of voltage and power factor with continuous values starting from 0 to
1. Therefore, the control signal derived from the de-fuzzification is a precise
instruction to the energy storage system to restore the voltage and power factor in a
relatively short period as compared to any basic control algorithm. They are
extremely undervoltage (EUV), very undervoltage (VUV), undervoltage (UV),
normal (Normal), overvoltage (OV), very overvoltage (VOV), and extremely
overvoltage (EOV) as shown in Figure 4.2. While the predefined fuzzy membership
for ΔPFy also includes seven different envelopes of power factor condition. Two of
69
the input parameters are not linked together. They will be triggered by different
condition and criteria. They are negative extremely high power factor (-EHPF),
negative high power factor (-HPF), negative power factor (-PF), undefined
(Undefined), positive power factor (+PF), positive high power factor (+HPF), and
positive extremely high power factor (+EHPF) as shown in Figure 4.3. These
memberships functions will be mapped to determine the types of instructions.
Figure 4.2: Fuzzy logic input variable membership for voltage change (ΔVy)
Figure 4.3: Fuzzy logic input variable membership for power factor change (ΔPFy)
70
For the defuzzification, there is also seven types of predefined membership functions,
namely extremely high supply power (EHSupply), high supply power (HSupply),
supply power (Supply), normal (Normal), absorb power (Absorb), high absorb power
(HAbsorb), and extremely high absorb power (EHAbsorb). The rules for the
defuzzification in terms of voltage change to be done are as shown in Table 4.1,
while Table 4.2 is due with the rules for defuzzification of power factor change. The
rules are designed in such a way to obtain the results oriented. After the membership
functions mapped each other, the centroid of defuzzification is used. The types and
the membership of the instructions are used to justify the area encircled in the fuzzy
output membership in Figure 4.4 and Figure 4.5. The changes needed for real current
(ΔIp) to minimize the unbalance voltage is sent as instruction to the bi-directional
inverter by using the center of area (CoA) deffuzification method. In the same way,
the change of reactive current (ΔIq) is needed to improve the power factor. The
linguistic variable ΔVy can vary in the range of -20 V to 20 V as the voltage
regulation is ±10% of the nominal value while the power factor can only vary from -
1 to 1.
Table 4.1: Rules of defuzzification for voltage change (ΔVy)
IF Delta Voltage = THEN Delta I =
EUV EHSupply
VUV HSupply
UV Supply
Normal Normal
OV Absorb
VOV HAbsorb
EOV EHAbsorb
71
Table 4.2: Rules of defuzzification for power factor change (ΔPFy)
IF Delta Voltage = THEN Delta I =
-EHPF EHSupply
-HPF HSupply
-PF Supply
Undefined Normal
+PF Absorb
+HPF HAbsorb
+EHPF EHAbsorb
Figure 4.4: Fuzzy output membership of instruction (ΔIp)
Figure 4.5: Fuzzy output membership of instruction (ΔIq)
72
For an example, if ΔVy is -6 V, it will generate two outputs:
1. 0.2 of very undervoltage (VUV)
2. 0.8 of undervoltage (UV)
Hence, the invoked rules for -6 V is "if delta voltage is VUV, then delta I is
Hsupply" and "if delta voltage is UV, then delta I is Supply" with the weight of 0.2
and 0.8 respectively. The instruction (ΔIp) sent to the bi-directional inverter can be
calculated through the CoA defuzzification method as shown in Figure 4.6.
Therefore, the output value will be -1.7. For the power factor, it is exactly similar to
the example above, if ΔPFy is -0.2, it will create two outputs:
1. 0.2 of undefined (Undefined)
2. 0.8 of negative power factor (-PF)
Therefore, the antecedents and consequences for -0.2 is "if delta power factor is
Undefined, then delta I is Normal" and "if delta power factor is -PF, then delta I is
Supply" with the weight of 0.2 and 0.8 respectively. The command (ΔIq) is delivered
after the calculation using the CoA defuzzification method and it is found to be -0.2
as shown in Figure 4.7.
73
Figure 4.6: CoA defuzzification method for real current change (ΔIp)
Figure 4.7: CoA defuzzification method for reactive current change (ΔIq)
4.2.2 Implementation of fuzzy Control Algorithm using LabVIEW
In LabVIEW, there is front panel and block diagram panel. Figure 4.8 shows the
developed graphical user interface of the energy storage system. Front panel is the
interface for the user to observe the changes of network parameters. From the front
panel, the user can monitor and control the network condition. It consists of the data
74
acquisition from the Data Taker with the basic parameters such as voltage, real and
reactive power, power factor, frequency, and current. The phase selection, VUF,
divergence, voltage sequence and other necessary parameters are also shown in the
interface for the purpose of inspection. Besides, the users can monitor and control the
three Sunny Islands through the interface manually or automatically. For manual
control, the user is allowed to control the amount of absorption and injection of
power in the Sunny Island. In automated mode, the fuzzy control algorithm will
trigger the energy storage system when the voltage is out of the acceptable range.
There are 15 buttons in the load bank controller to control the emulated consumer
load. Graphs will demonstrate the dynamic change of voltage, VUF and power factor.
Figure 4.8: Graphical User Interface of the Energy Storage System
The block diagram contains a lot of wired functional blocks. They are divided into a
few parts, which are the data acquisition section, fuzzy control section, computing
section, control of Sunny Island, and others. The details of the parts will be further
elaborated in the following paragraphs. Figure 4.9 shows a simple load bank control
75
where it controlled by the 15 buttons shown on the monitoring interface in Figure 4.8.
The 15 buttons are clustered as a whole in order to arrange them properly.
Figure 4.9: Load bank control
4.2.2.1 Coding of the Data Acquisition
The functional blocks used to read the measurement data from the Data Taker is
shown in Figure 4.10. In order to establish a working connection, the TCP open
connection with the IP address 169.254.60.152 is used to connect the Data Taker and
read the input registers of the 5 power meters of PV1, PV2, load bank, generator, and
wind turbine. Since the data read from the registers are encoded values, all the data
are decoded as shown in Figure 4.11 in order to obtain the parameters correctly.
Figure 4.10: Values read from the input registers
76
Figure 4.11: Way to decode the encoded the parameters obtained
4.2.2.2 Saving of the data collected
After all the data acquisition, data obtained will be saved in TDMS format in the
desired location. For TDMS format file, it can be generated in the template excel file
with all the group names (sheet) and channel names (column) are given. Hence, the
data will be grouped and displayed in different sheets as desired. The user can review
and analyse the data loaded from the file directly as shown in Figure 4.12. Figure
4.13 shows the coding for storing of the data. The data measured by the Sunny
77
Islands and power meters will be saved accordingly. The specified TDMS file will be
created or replaced to the specific file path.
Figure 4.12: Data saved in excel format
Figure 4.13: Data Collection coding
78
4.2.2.3 Control of the Sunny Island
Figure 4.14 illustrates the way to turn ON and OFF the Sunny Island with either
signal ‘1’ or ‘2’ is sent. Once the signal is sent to turn ON, IO server variable pre-set
in the IO server library will connect with the YAOPC server which is under the SMA
possession protocol. Next, true and false selector is used by the controller to identify
whether the instruction sent to the Sunny Island is taken from the automated control
or manual control as shown in Figure 4.15. When the instruction transmitted is the
same as the previous iteration, it flows into the false case and no action is taken in
the Sunny Island. This can prevent the communication port from doing the extra job
and avoiding the delayed problem. As shown in Figure 4.16, the parameters of the
Sunny Island are read from the IO server variable. Batteries of each phases within a
range of 60% - 90% is set as healthy condition. Power output of the Sunny Island is
plotted to have a better view on monitoring.
Figure 4.14: Turning ON and OFF of the Sunny Island
79
Figure 4.15: Controlling of Sunny Island
Figure 4.16: Reading parameters from Sunny Island
4.2.2.4 Coding of the Main Fuzzy Controller
For the programming of the main fuzzy controller, it is put into a sequence structure.
Firstly, the voltage and current are obtained by DAQ Assistant and the fundamental
of voltage and current signals for N channels are calculated. By using the complex to
polar function, the output received will be converted to the magnitude and phase
components. Hence, the voltage and current magnitude can be obtained while the
voltage angle and the current angle can be calculated from the phase component.
80
Therefore, VUF can be calculated when the negative sequence is divided by positive
sequence and times 100. Figure 4.17 shows the block diagram for the calculation of
VUF.
Figure 4.17: Calculation of VUF
By utilizing the symmetrical components, maximum divergence can be determined
and phase with the maximum divergence will be corrected by the energy storage
system. Park’s Transformation as shown in Figure 4.18 will be adopted as the phase
selection for the fuzzy controller. Park’s transformation is used to determine the
maximum divergence of phase, while the phase selection for low power factor is
shown in Figure 4.19. The condition to trigger the case is when the power factor is
lower than 0.98. The phase with the lowest power factor will be selected as the
severe phase. However, when the voltage unbalance and the power factor are within
the tolerance, it will go to the default case, hence, the fuzzy controller will maintain
the previous instruction with no increment or deduction of current output towards the
Sunny Islands. Figure 4.20 and Figure 4.21 show the fuzzy logic for voltage
unbalance and power factor used respectively. It is explained in the section 4.2.1
81
which describes the details of the fuzzy controller. The output of the fuzzy logic is
the current value that needed by the energy storage system for adjustment of the
power output.
Figure 4.18: Park’s Transformation
Figure 4.19: Phase selection for low power factor
82
Figure 4.20: Fuzzy logic for voltage unbalance
Figure 4.21: Fuzzy logic for power factor
4.2.2.5 Network Operating Conditions Causing the Operation of the Energy
Storage System
For the energy storage system to supply or absorb power from the networks, the
following situations must be fulfilled. First, the voltage must be greater than 230 V
and updated in the following iteration for the case to be triggered. The case will
check the active current value whether it is within the defined safety range which is -
16 A to 16 A. Another condition that needs to be fulfilled is that the voltage of the
distribution network must be higher than 1 which is set in the controller in order to
83
make sure that the network is always energised; otherwise the energy storage system
can continue to operate on the dead network. Then, the present current value will
subtract the output current specified by the fuzzy logic. If the current value after
subtraction is equals to 0 A, the case will check for the present voltage whether it is
smaller than the average voltage. If it is smaller than the average voltage, the current
will be added by 1 A. In vice versa, current will become -1 A. This is to prevent the
energy storage system to stop working. Finally, the current value will be sent to the
energy storage system in order to mitigate the voltage unbalance of the network.
Figure 4.22 shows the cases as mentioned above.
Figure 4.22: Condition of active current before sent to the energy storage system
On the other hand, the conditions for reactive current are slightly different. First, the
reactive current is checked to be within the range of -16 A to 16 A. The acceptable
voltage range for the case is within 238 V to 243 V. If the power factor is in the
range of 0.3 to 0.98, the present current value will be added by the current output of
the fuzzy logic. However, if the power factor is in the range of -0.3 to -0.98, the
output of fuzzy logic will be subtracted by the current value. After all, the reactive
current instruction will be sent to the energy storage system to improve the power
84
factor of the network. Figure 4.23 shows the situation and the way for the reactive
current to process before sent to the energy storage system.
Figure 4.23: Processing of reactive current before sent to the energy storage system
After all, the energy storage system will operate by adjusting the power output from
the bi-directional inverter after the amount of the real and reactive current needed to
inject or absorb had been determined. Hence, voltage rise, voltage unbalance, and
low power factor can be mitigated by this fuzzy logic controlled energy storage
system effectively.
4.3 Summary
Fluctuations of RE output and random plugged-in EVs are unpredictable, causing
voltage rises, voltage sags, voltage unbalance, and low power factor. Therefore, an
effective active control algorithm is needed to mitigate the fluctuating voltage rises,
voltage unbalance and improve power factor. This chapter presents a fuzzy control
method implemented with the energy storage system. The fuzzy control algorithm is
85
developed using LabVIEW in a supervisory personal computer. The fuzzy control
would determine and send the appropriate instruction to the energy storage system
and instruct the energy storage system to carry out the necessary action to prevent the
network parameters exceed their statutory limits. The Park’s transformation is used
in the algorithm to identify the affected phase that causing high degree of voltage
unbalance factor and hence to coordinate with the three-phase energy storage system
to mitigate the voltage rise, voltage sags and voltage unbalance issues. Hence, the
particular affected phase can be addressed and balance the voltage effectively.
Meanwhile, only the lowest power factor will be addressed for each phases to
improve the power factor.
86
CHAPTER 5
RESULTS AND DISCUSSIONS
5.1 Introduction
The performance of the fuzzy controlled energy storage system is verified by setting
up the energy storage system on the experimental LV distribution network with two
single-phase PV systems, a wind turbine emulation system, and an electric vehicle. A
number of case studies under different generations load demands are designed to
investigate the performance of the three single-phase fuzzy controlled energy storage
systems. The discussion of the results and conclusions are presented in this chapter.
5.2 Experimental Result and Discussion
The renewable energy used in this system is the 3.6 kWp PV system and 2.3kWp
wind emulation system. When the RE is connected to the network, it causes VUF to
fluctuate and increase. The statutory limit of VUF set by Malaysian Grid Code is 1 %.
Hence, the fuzzy logic controlled energy storage system is designed to react in such a
way to prevent the network VUF from exceeding 1 %. In addition, penalty will be
charged when power factor is lower than 0.85. Hence, this energy storage system is
87
also designed in such a way to improve the power factor. Several case studies are
conducted to represent the different scenarios.
5.2.1 Case Study 1: Impacts of high PV power output on the experimental LV
distribution network
The idea of this case study is to investigate the impact of the intermittent PV power
output on the experimental network. A constant load of 700 W and a 3.6 kWp PV
system are connected at phase A of the experimental network. Figure 5.1 shows the
load profile, PV power output and the power factor of the experimental case study.
Figure 5.2 shows the voltage and VUF at the PCC.
Figure 5.1: PV power output at Phase A, load at Phase A and power factor of all
three phases of the experimental case study 1
88
Figure 5.2: VUF and voltage at phase A of the experimental case study 1
Initially, the PV system generates 800 W which is slightly greater than the 700 W
load. The VUF is recorded at 1% while the voltage is approximately 243 V. However,
at 2:45 pm, the PV power output starts to increase, causing the system voltage to rise
above 251 V and the VUF to be 3.4%. It is noticed that the power factor is
fluctuating all the time. This is due to the intermittent PV power output.
5.2.2 Case study 2: Power factor of a university’s building
The second case study is carried out to study the power factor profile at the tutorial
block (SE Block) in the university. This data was collected on the 1st July 2013 from
12 am to 11.59 pm. This one day data was taken at the switch room of SE Block
using TES-3600 3-phase power analyzer.
89
Figure 5.3 is the power factor profile on Monday, a working day. It is shown that the
power factor is in the range of 0.75 and 0.85 early in the morning. However, the
power factor begins to increase to the range of 0.8 and 0.9 starting at 8 am. This is
because the occupants started to switch on many electronic devices in SE Block such
as lighting, air conditioner, fans, computers and even some of the laboratory
equipment.
Figure 5.3: Power factor profile on 1st July 2013, Monday
Figure 5.4 shows the power factor profile on a non-working day, Saturday. It
is shown that the power factor stays within the range of 0.75 and 0.85 throughout the
day. This is because the occupants didn’t switch on many electronic devices. Case
study 2 shows that the power factor at the building of the university is always low.
However, if the energy storage system is used, the power factor of the building is
improved.
90
Figure 5.4: Power factor on 6th July 2013, Saturday
5.2.3 Case Study 3: Effect of using fuzzy controlled energy storage on the
network with PV
This case study is to study the response of the fuzzy controlled energy storage system
with respect to the changes of PV power output under a balanced load condition. A
fixed load of 400 W is introduced to every phases. A 3.6 kWp PV system is
connected to phase B of the network.
Figure 5.5 shows that, the increment of PV power output affects the three-phase
voltages at 3.38 pm. Due to the coupling effect, voltage at phase C reduces to 230 V
although the PV exports power to phase B. The fuzzy control system detects the
sudden change and instructs the bi-directional inverter to absorb the excess power
from the PV system at phase B. The three-phase voltages are then restored to their
nominal values once the excess power is stored into the energy storage system. At
91
3.45 pm, the control algorithm detects that the voltage at phase C is out of range.
Therefore, the controller instructs the energy storage system to restore the voltage.
Figure 5.5: Three-phase voltages and PV power output of the experimental
case study 3
Figure 5.6 shows the response of the energy storage system and the VUF of the
experimental network. The VUF is recorded at 2% when the PV system exports
power to the network. It is noticed that VUF is reduced to 0.5% in 1 minute when the
energy storage system absorbs the excess power from the network. As compared to
the results of (Wong, Lim, & Morris, 2014) in Figure 5.7, VUF is found mitigated from
1.9% to 0.9% in 5 minutes. Table 5.1 shows the improvement of power quality with the
integration of the fuzzy controlled energy storage system.
92
Figure 5.6: VUF and power output of the energy storage system of the experimental
case study 3
Figure 5.7: Corresponding VUF of the energy storage system using the control
strategy of (Wong, Lim, & Morris, 2014)
93
Table 5.1: Comparison of the improvement of VUF with the integration of the
energy storage system
Methods VUF
Before
VUF
After
VUF
Improvement
Time
Needed
Fuzzy Control
(Wong, Lim, &
Morris, 2014)
1.9 0.9 52.6% 5 mins
My method 2.0 0.5 75% 1 mins
Figure 5.8 shows that the power factor of the network. There is a drop in power
factor at 3.38 pm because the PV inverter begins to be connected to the network,
exporting real power to the network at that time and hence causing a reduction on the
power factor. The reduction in the power factor happens only for a brief period of
time after the connection of the PV system at phase B at 3.37 pm. The power factor
is restored automatically to the normal level without any corrective effort from the
energy storage system because the PV inverter has the built-in feature to bring up the
power factor to 0.85 or above at its terminal once it is connected. However, the
power factor can still be easily reduced if any electrical vehicles or electrical
machines are plugged to the network. Therefore, the energy storage has to be used in
conjunction with the PV inverter so that the power factor can be increased up to 1.00.
94
Figure 5.8: Power factor of all three-phases and reactive power output of the
energy storage system
5.2.4 Case Study 4: Effect of using energy storage when inductive load is
connected.
This case study is to investigate the effectiveness of the fuzzy controlled energy
storage system in improving the power factor of the experimental network. In this
experiment, a 2600 mH inductive load is connected to phase A. Figure 5.9 shows the
VUF, voltage and its angle, power factor, reactive current control of the energy
storage system and reactive power flow of the experimental network. It is noticed
that VUF is reduced from 0.8% to 0.5% after the energy storage system supplies 192
Var to compensate the inductive load in the experimental network. The fuzzy control
algorithm is used to manipulate the amount of reactive current flow from the energy
storage system to the network based on the network power factor. It is noticed that
the power factor is maintained within 0.98 and unity power factor. Table 5.2 shows
the improvement of power factor and VUF with the integration of the fuzzy
95
controlled energy storage in this case study. Reactive power is not only able to
improve the power factor; it can also provide a support to the network voltage. This
indicates that reactive power supplied by the energy storage system can improve
power factor, VUF, and voltage.
Figure 5.9: (a) Three-phase voltages, (b) VUF, (c) Reactive power flow and,
(d) Power factor of the experimental case study 4
Table 5.2: Improvement of power qualities with the integration of the fuzzy
controlled energy storage system
Power Quality Before After Improvement
Power Factor 0.93 0.98 5%
VUF 0.8 0.5 37.5%
96
5.2.5 Case study 5: Effect of using energy storage when electric car (EV) and
photovoltaic system (PV) are connected to the network
This case study is to study the response of the proposed energy storage system when
EV and PV system are connected at the same time. Figure 5.10 shows the VUF, EV
power, PV power output and voltage at phase A. At 3.16 pm, EV is charged at 2.4
kW, hence causing the voltage at phase A to drop below 230 V. It is noticed that the
fuzzy controller detects the sudden change and instructs the energy storage system to
inject power to the network. At 3.19 pm, PV is connected to phase A of the network,
hence causing the voltage at phase A to increase. It is noticed that the fuzzy control
algorithm is able to manipulate the power output of the energy storage system so to
mitigate the VUF and voltage of the experimental network. At 3.25 pm, the PV and
EV are disconnected on a purpose and it is shown that the fuzzy controller is able to
regulate the voltage and mitigate the VUF in a very short period.
Figure 5.11 shows the power factor at Phase A, B and C with the reactive current
from ESS. The power factor at phase A starts to drop at 3.17 pm because EV begins
to draw reactive power only from the grid emulator after ESS supplies real power to
EV at 3.17 pm to improve the voltage level. However, the fuzzy controller detects
the reduction of the power factor and then supplies reactive power to the network to
improve the power factor. At 3.19 pm, the power factor at Phase A drops again. This
is because the PV starts to supply real power to the load at 3.19 pm, hence causing
the power factor to drop. However, the reduction happens for a brief period of time
because the inverter has the built-in feature to correct the power factor at its terminal.
At 3.25 pm, the power factor drops again after PV and EV are disconnected from the
97
network. However, the fuzzy controller senses the reduction and then supplies
reactive power to the network to restore the power factor.
Figure 5.10: VUF, voltage, active power output of energy storage system, and PV
power of the experimental case study 5i
Figure 5.11: Power factors of all three phases and reactive current output from the
energy storage system
98
5.2.6 Case Study 6: Impacts of wind power on the experimental LV
distribution network
In this case study, emulated wind energy is injected in order to study its impact to the
experimental LV distribution network. The wind speed is emulated based on the
wind speed graph of Kuala Terengganu for the year of 1995 in Figure 5.12
(Shamshad Ahmad, 2009). The graph shows that the wind speed is fluctuating. A
constant load of 370 W is connected to each of the phases while wind emulator is
connected to phase A of the experimental network. This wind emulator is designed to
fluctuate along the time in order to emulate the harvested wind energy. However, the
wind power that can be generated is around 600 W only. Figure 5.13 shows the wind
power output and the power factor of the experimental case study.
Figure 5.12: Mean daily wind speed values at Kuala Terengganu (Shamshad Ahmad,
2009)
99
Figure 5.13: Wind power output at phase A and power factor of three-phases in
experimental case study 6
When wind power output is fluctuating, the power factor of phase A is also
fluctuating according to the wind power. Due to the network is highly loaded with
inductance load; power factor is only 0.8 for all the three phases. It is noticed that the
power factor drops tremendously during high wind power output. This is because the
induction machine is absorbing the reactive power from the network.
Figure 5.14 shows the VUF and voltages of three-phases of the experimental network.
It can be noticed that VUF and voltages are affected by the fluctuating wind power
output. Due to small wind power output, VUF hits only 0.9% which is still under the
statutory limit. However, when this situation is applied to the higher wind power
output, VUF and voltages can be out of the range hence creating power quality issues.
100
Figure 5.7: VUF and three-phase voltages of the experimental case study 6
5.2.7 Case Study 7: Effect of using fuzzy controlled energy storage on the
network with wind emulator.
This case study is to study the response of the fuzzy controlled energy storage system
with respect to the changes of wind power output under a balanced load condition. A
fixed load of 370 W is connected to every phase and wind emulator is connected to
phase A of the network.
Figure 5.15 shows the fluctuating of wind power output affecting the three-phase
voltages. Figure 5.16 shows the response of the energy storage system and the VUF
of the experimental network. Since the wind power output is around 600 W only,
VUF hardly to reach the statutory limit of 1%. For the fuzzy logic control, if VUF
hits 0.9%, the energy storage system will react to reduce the hazard of high VUF. At
11.06 am, it can be noticed that, the energy storage system responses to the high
VUF.
101
Figure 5.8: Three-phase voltages and wind power output of the experimental
case study 7
Figure 5.9: VUF and power output of the energy storage system of the experimental
case study 7
Figure 5.17 shows the response of the energy storage system and the power factor of
the experimental network. Since the inductive load is connected to the network, the
power factor is around 0.9. It can be observed that when the energy storage system is
used to improve the power factor, the power factor is improved to unity. However,
102
the wind emulator exporting power to the network causes the power factor to drop. It
is because the induction machine absorbs the reactive power from the network. The
fuzzy logic control then injects reactive power to the network to improve the power
factor.
Figure 5.10: Power factor of all three-phases and reactive power output of the
energy storage system for experimental case study 7
5.3 Summary
Several scenarios under different generations and loading conditions were outlined in
this chapter. The performance of the fuzzy controlled energy storage is verified by
setting up the energy storage system on the experimental LV distribution network.
The responses of the fuzzy control system under different conditions were
investigated. The experimental results show that the fuzzy controlled energy storage
system is able to restore the voltage fluctuation, reduce the voltage unbalance factor,
and improve the low power factor under rapid changes of RE output and EV
integration. Therefore, the developed control algorithm is found satisfy to limit the
103
desired network parameters within the level under intermittent RE and integration of
EV. At the same time, the fuzzy controlled energy storage can increase the use of the
RE without limiting the generation effectively.
104
CHAPTER 6
CONCLUSION
6.1 Conclusion
The utilities of the world are taking solid steps towards incorporating new
technologies to evolve the traditional grid systems into smart grid systems. As a
promising renewable alternative, the wind and solar power is highly expected to
contribute a significant part of generation in power systems in the future, but this also
brings new integration related power quality issues. In line with this, this thesis
presents a fuzzy controlled energy storage system to mitigate the VUF, voltage rise,
and improve the power factor by manipulating the real and reactive power flow.
Several scenarios under various generating and loading conditions have been carried
out to investigate the performance of the fuzzy controlled energy storage system. It is
noticed that the energy storage system which allows real power control only has its
limited capability. With the aid of reactive power in the distribution network, it can
provide voltage support and also improve power factor. The experimental results
show that the proposed fuzzy controlled energy storage system not only mitigates the
105
voltage rise and VUF caused by intermittent PV power output, but also improves the
network power factor effectively.
The majority of the existing control methods are used to solve steady state voltage
fluctuation, voltage unbalances and power factor. The fuzzy control method is
different from the existing control methods as it is able to adjust the supply of its real
and reactive power appropriately and effectively based on the changes in voltage
magnitude and power factor at a point of concern. As a result, the voltage magnitude
and power factor can be maintained well within the required tolerance under the high
intermittency of photovoltaic systems which happened predominantly in Southeast
Asia or other regions with cloudy skies. In the developed control algorithm, VUF
can be reduced from 2% to 0.5%, which is a 75% of reduction in 1 minute. Power
factor is also improved from 0.93 to 0.98. Therefore, this method is an appropriate
choice when dealing with dynamic voltage fluctuations, voltage unbalances and
power factor. Also, this fuzzy control solution is definitely a new and powerful
approach for the energy storage system.
6.2 Future Works
Further work is necessary to study how the energy storage system can solve power
quality issues caused by the random penetration of EV, ESS and PV across the real
distribution network. Also extra work is required to establish the appropriate size of
the energy storage system with respect to the capacity of the RE systems. This is
106
because the cost of the energy storage system is also an important aspect to be
considered for the implementation.
PUBLICATIONS
Journal Paper Published:
Lim, K. Y., Lim, Y. S., Wong, J. & Chua, K. H., 2015. Distributed Energy Storage
with Real and Reactive Power Controller for Power Quality Issues Caused by
Renewable Energy and Electric Vehicles. ASCE Journal of Energy Engineering.
Conference Paper Published:
Lim, K. Y., Lim, Y. S., Wong, J. & Chua, K. H., 2014. Energy Storage with Fuzzy
Controller for Power Quality Improvement on Distribution Network on Distribution
Network with Renewable Energy. Tunisia, Setcor.
107
REFERENCES
Aarathi A.R, D. M. J., 2014. Grid Connected Photovoltaic System with Super
Capacitor Energy Storage and STATCOM for Power System Stability Enhancement.
Thiruvananthapuram, IEEE, pp. 26-32.
Abba-Aliyu, S., 2009. Voltage Stability and Distance Protection Zone3, Göteborg,
Sweden: CHALMERS UNIVERSITY OF TECHNOLOGY .
Abdul Rahman Mohamed, K. T. L., 2006. Energy for sustainable development in
Malaysia: Energy policy and alternative energy. ELSEVIER, pp. 2388-2397.
Alvarez, R. A. et al., 2012. Greater focus needed on methane leakage from natural
gas infrastructure. s.l., s.n., p. 6435–6440.
Andreas Poulikkas, S. P. G. K. I. H., 2014. Storage Solutions for Power Quality in
Cyprus' Electricity Distribution Network. AIMS's Energy, Volume 2, Issue 1, DOI:
10.3934/energy.2014.1.1, pp. 1-17.
Anon., 2005. Overview of Policy Instruments for the Promotion of Renewable
Energy and Energy Efficiency in Malaysia. [Online]
Available at:
http://1solar.ecosensa.com/images/Overview%20of%20Policy%20for%20the%20Pr
omotion%20of%20Renewable%20Energy%20and%20Energy%20Efficiency%20in
%20Malaysia.pdf
Anon., 2014. Local Automotive News. [Online]
Available at:
http://mai.org.my/v3/index.php?option=com_k2&view=item&id=490:rm3-million-
fund-set-up-for-300-ev-charging-stations
[Accessed 16 December 2014].
108
Bajo, C. G. et al., 2015. Voltage Unbalance Mitigation in LV Networks Using Three-
Phase PV Systems. s.l., IEEE, pp. 1-5.
Bishop, T. H., 2008. Unbalanced Voltages and Electric Motors. [Online]
Available at:
http://www.industrialekg.com/Library/Technical%20Papers%20by%20Third%20Par
ties/unbalanced-voltages-and-electric-motors.html
Bollen, M., 2002. Definitions of Voltage Unbalance. Power Engineering Review,
IEEE, pp. 49-50.
Buchmann, I., 2009. Non-correctable Battery Problems. [Online]
Available at:
http://batteryuniversity.com/learn/article/non_correctable_battery_problems
[Accessed 2013].
Calstart, 2010. ALLCARINDEX. [Online]
Available at: http://www.allcarindex.com/auto-car-model/United-States-Calstart-
Showcase-Electric-Vehicle-SEV/
[Accessed 2 September 2014].
Canada, M. o. N. R., 2004. Final Report on the August 14, 2003 Blackout in the
United States and Canada: Causes and Recommendations, Canada: U.S.-Canada
Power System Outage Task Force.
Castillo, M. D., Jr., G. P. L., Yoon, Y. & Chang, B., 2014. Application of Frequency
Regulation Control on the 4MW/8MWh Battery Energy Storage System (BESS) in
Jeju Island, Republic of Korea. Journal of Energy and Power Sources, 1(6), pp. 287-
295.
Chen, T.-H., Yang, C.-H. & Hsieh, T.-Y., 2009. Case Studies of the Impact of
Voltage Imbalance on Power Distribution Systems and Equipment. United States,
WSEAS, pp. 461-465.
109
Chua, K. H. et al., 2012. Energy Storage System for Voltage Unbalance Mitigation in
Low Voltage Distribution Networks with Photovoltaic Systems. IEEE Transactions
on Power Delivery, October, 27(No.4), pp. 1783-1790.
Chua, K. H. et al., 2012. Voltage Unbalance Mitigation in Low Voltage Distribution
Networks with Photovoltaic Systems. Journal of Electronic Science and Technology,
March.10(No 1).
Chuang, A. S. & Gellings, C., 2008. Demand-side Integration in a Restructured
Electric Power Industry. France, Electric Power Research Institute, pp. 1-9.
Dabur, P., Singh, G. & Yadav, N. K., 2012. Electricity Demand Side
Management:Various Concept and Prospects. International Journal of Recent
Technology and Engineering (IJRTE), pp. 1-6.
Divya, K. & Østergaard, J., 2008. Battery energy storage technology for power
systems—An overview. Electric Power Systems Research, 79(4), p. 511–520.
E. Ghahremani, I. K., 2014. Optimal Allocation of STATCOM with Energy Storage
to Improve Power System Performance. Chigago, IL, USA, IEEE, pp. 1-5.
E.A. Mertens Jr, L. F. B. J. H., 2007. Evaluation and tends of power quality indices
in distribution system. Barcelona, IEEE, pp. 1-6.
Farhoodnea, M., Mohamed, A., Shareef, H. & Zayandehroodi, H., 2013. Power
quality impacts of high-penetration electric vehicle stations and renewable energy-
based generators on power distribution systems. Measurement, Volume 46, Issue 8,
October 2013, pp. 2423-2434.
Feldman, D. & Barbose, G., 2014. Photovoltaic (PV) Pricing Trends: Historical,
Recent, and Near-Term Projections (2014 Edition), United State: National
Renewable Energy Laboratory (NREL).
110
Ferry A. Viawan, S. M. I. D. K. S. M. I., July 2008. Coordinated Voltage and
Reactive Power Control in the presence of distributed generation. Pittsburgh, PA,
IEEE, pp. 1-6.
Forster, F., 2010. EE Times Europe Automotive. [Online]
Available at: http://www.automotive-eetimes.com/en/digital-isolation-in-hybrid-and-
electric-vehicles.html?cmp_id=71&news_id=222901126&page=0
[Accessed 17 September 2015].
Ghavameddin Nourbakhsh, B. T. G. M. A. G., 2013. Distribution Tap Changer
Adjustment t Improve Small-Scale Embedded Generaor Penetration and Mitigation
Voltage Rise. Hobart, TAS, IEEE, p. 5.
Gosbell, V. J., 2002. Voltage Unbalance. [Online]
Available at: http://www.elec.uow.edu.au/apqrc/content/technotes/technote6.pdf
GreenTech, M., 2013. Standards for Electric Vehicles. [Online]
Available at:
http://www.greentechmalaysia.my/content.asp?zoneid=2&articleid=243&cmscategor
yid=407#.VcLeafmqqko
H.M.S.Chandana Herath, S. M., 2008. Power Quality Trends in Energy Australia
Distribution Network. Wollongong,NSW, IEEE, pp. 1-6.
Hendri Masdi, N. M. S. M. A. M. a. S. Y., 2004. Design of a Prototype D-Statcom
for Voltage Sag Mitigation. Malaysia, IEEE, pp. 61-66.
Hoppecke, 2012. Sizing of lead acid batteries for solar application, Germany:
Hoppecke.
IEEE, 2005. IEEE Recommended Practice for Powering and Grounding Electronic
Equipment. United States of America: Institute of Electrical and Electronics
Engineers, Inc..
111
International Monetary Fund, 2016. Commodity Prices of Natural Gas. [Online]
Available at: http://www.indexmundi.com/commodities/?commodity=natural-
gas&months=60¤cy=myr
International Renewable Energy Agency (IRENA), 2015. Battery Storage for
Renewables: Market Status and Technology Outlook, s.l.: International Renewable
Energy Agency (IRENA).
Jenkins, N. e. a., 2000. Embedded Generation. London: IET.
Jiménez, A. & García, N., 2012. Unbalanced three-phase power flow studies of
distribution systems with plug-in electric vehicles. North American, IEEE, pp. 1-6.
Johnson, K., 2007. The Fundamentals of Linking Demand Side Management
Strategies with Program Implementation Tactics, Strategies Newsletter: Association
of Energy Services Professional.
KeeGan Dorai, L. K. B. A., 2012. Motor Trader. [Online]
Available at: http://www.motortrader.com.my/news/first-public-ev-charging-
stations-in-malaysia/
[Accessed 22 September 2015].
Kenneth Gillingham, R. G. N. a. K. P., 2004. Retrospective Examination of Demand-
Side Energy Efficiency Policies, Washington: Resources for the future.
Kottick, D., Res. & Dev. Div., I. E. C. L. H. I., Blau, M. & Edelstein, D., 2002.
Battery energy storage for frequency regulation in an island power system. IEEE
Transactions on Energy Conversion (Volume:8 , Issue: 3 ), 8(3), pp. 455 - 459.
Lee, C.-Y., 1999. Effects of unbalanced voltage on the operation performance of a
three-phase induction motor. IEEE Transaction Energy Conversion, pp. 202-208.
112
Liang, L., Zhong, J. & Jiao, Z., 2012. Frequency Regulation for a Power System with
Wind Power and Battery Energy Storage. Auckland, IEEE, pp. 1-6.
Li, F. et al., 2008. A Framework to Quantify the Economic Benefit from Local VAR
Compensation. International Review of Electrical Engineering (IREE), pp. 989-998.
Li, K., Liu, J., Wang, Z. & Wei, B., 2007. Strategies and Operating Point
Optimization of STATCOM Control for Voltage Unbalance Mitigation in Three-
Phase Three-Wire Systems. Power Delivery, IEEE Transactions, Volume:22, Issue:1,
pp. 413-422.
Lim Yun Seng, P. T., 2006. Innovative Application of Demand Side Management to
Power Systems. Sri Lanka, IEEE Conference Publications, pp. 185-189.
Lim, Chia Ying, 2015. Electric vehicles are the face of future mobility. Malaysia:
Star Media Group Berhad.
Lim, Y. S., WHITE, S., NICHOLSON, G. & TAYLOR, P., 6-9 June 2005.
Additional applications of demand side management techniques in power systems
integrated with distributed generation. Turin, Italy, IEEE, pp. 1-5.
Lu, C.-F., Liu, C.-C. & Wu, C.-J., 2002. Effect of battery energy storage system on
load frequency control considering governor deadband and generation rate constraint.
IEEE Transactions on Energy Conversion (Volume:10 , Issue: 3 ), 10(3), pp. 555 -
561.
Lyons, P. F., Trichakis, P., Taylor, P. C. & Coates, G., 2010. A Practical
Implementation of A Distributed Control Approach for Microgrids. Intelligent
Automation & Soft Computing, pp. 319-334.
Malaysia Building Integrated Photovoltaic, M., 2010. PV System Cost. [Online]
Available at: http://www.mbipv.net.my/content.asp?zoneid=4&categoryid=12
[Accessed 27 August 2012].
113
Marcos, J., Marroyo, L., Lorenzo, E. & García, M., 2012. Power output fluctuations
in large PV plants. Santiago de Compostela (Spain), European Association for the
Development of Renewable Energies, Environment and Power Quality.
Mat Siddique, G. a. B. S., 2004. Effects of Voltage Unbalance on Induction Motors.
s.l., IEEE, pp. 26-29.
Mohibullah, S. H. L., 2012. Power Quality Issues and Need of Intelligent PQ
Monitoring in the Smart Grid Environment. London, IEEE, pp. 1-6.
Moosavian, S. R. N. S. J. S. K., 2013. Energy policy to promote photovoltaic
generation. Renewable and Sustainable Energy Reviews, pp. 44-58.
Muhammad A. Saqib, A. Z. S., 2015. Power-quality issues and the need for reactive-
power compensation in the grid integration of wind power. Renewable and
Sustainable Energy Reviews 43, pp. 51-64.
Natgas, 2013. Natural Gas and the Environment. [Online]
Available at: http://naturalgas.org/environment/naturalgas/
Networks, F. E., 2015. First Energy Networks. [Online]
Available at: http://www.firstenergy.com.my/charger-locations/search-charger-
locations
[Accessed 14 September 2015].
Ning, Z., Luis, F. & Daniel, S. K., 5-7 December 2011. Investigating the impact of
demand side management on residential customers. Manchester, IEEE, pp. 1-6.
Oeng, L. & Sangwongwanich, S., 2016. A Voltage Rise Mitigation Strategy under
Voltage Unbalance for a Grid-connected Photovoltaic System. Chiang Mai, Thailand,
Elsevier, pp. 209-312.
114
Pillay, P. & Manyage, M., 2001. Definitions of Voltage Unbalance. IEEE Power
Engineering Review, pp. 50-51.
Prakash K. Ray, S. R. M. a. N. K., 2013. Classification of Power Quality
Disturbances Due to Environmental Characteristics in Distributed Generation System.
Sustainable Energy, IEEE Transactions on (Volume:4 , Issue: 2 ), pp. 302 - 313.
Price, A., Bartley, S., Male, S. & Cooley, G., 1999. A Novel Approach to Utility
Scale Energy Storage. Power Engineering Journal, pp. 122-129.
Qiushuo Li, S. T. X. X. J. W., 2013. Monitoring And Analysis of Power Quality in
Electric Vehicle Charging Stations. Tainan, IEEE, pp. 277-282.
R. Romo, O. M., 2015. Power quality of actual grids with plug-in electric vehicles in
presence of renewables and micro-grids. Renewable and Sustainable Energy Reviews,
pp. 189-200.
R. Sastry Vedam, M. S. S., 2008. Power Quality: VAR Compensation in Power
Systems. London: CRC Press .
REEEP, R. E. &. E. E. P., 2006. Sustainable Energy Regulation and Policy making
Training Manual [pdf]. [Online]
Available at: http://africa-toolkit.reeep.org/modules/Module14.pdf
[Accessed 12 May 2012].
Rjakaruna, E. b. S., Shania, F. & Ghosh, A., 2015. Plug In Electric Vehicles in Smart
Grids: Integration Techniques. Singapore: Springer.
Roberto Caldon, M. C. R. T., 2012. Voltage Unbalance Compensation In LV
Networks With Inverter Interfaced Distributed Energy Resources. University of
Padova, Italy, IEEE, pp. 527-532.
115
Ryan Eastman, S. G. W. C. J. H., 2014. Climatic Atlas of Clouds Over Land and
Ocean. [Online]
Available at: http://www.atmos.washington.edu/CloudMap/
[Accessed 18 September 2015].
Ryan Eastman, S. G. W. C. J. H., 2014. Total Cloud Cover:Completely Clear
Sky:Frequency of Occurance. [Online]
Available at: http://www.atmos.washington.edu/CloudMap/
S.Mekhilef, D., 2011. Assessment of off-shore wind farms in Malaysia. Bali, IEEE,
pp. 1351-1355.
Salman Habib, M. K. U. R., 2015. Impact analysis of vehicle-to-grid technology and
charging strategies of electric vehicles on distribution networks – A review. Journal
of Power Sources, pp. 205-214.
Schoenung, S. M., 1996. Energy Storage For a Competitive Power Market. Energy
and the Environment, pp. 347-370.
Sergio Faias, J. S. R. C., 2009. Embedded Energy Storage Systems in the Power Grid
for Renewable Energy Sources Integration. In: Renewable Energy. Portugal: InTech,
pp. 63-88.
Shahnia, F., Ghosh, A., Ledwich, G. & Zare, F., 2010. Voltage Unbalance Reduction
in Low Voltage Distribution Networks with Rooftop PVs. Australia, IEEE, pp. 1-5.
Shamshad Ahmad, W. H. W. B. M. M. S. S., 2009. Open Access Repository of USM
Research & Publication. [Online]
Available at:
http://eprints.usm.my/10959/1/Analysis_of_Wind_Speed_Variations_and_Estimatio
n_of_Weibull_Parameters_for_Wind_Power_(PPK_Awam).pdf
[Accessed 22 October 2015].
116
Sharad.W.Mohod, D. M. V., 2008. Power Quality Issues & It's Mitigation Technique
in Wind Energy Generation. Wollongong, NSW, IEEE, pp. 1-6.
Singh, B. R. & Singh, O., 2010. 21st Century challenges of clean energy and global
warming-can energy storage systems meet these issues?. Cairo, IEEE, pp. 323 - 329.
Sopian, K., Haris, A., Rouss, D. & Yusof, M. A., 2005. Building Integrated
Photovoltaic (BIPV) in Malaysia- Potential, Current Status Strategies For Long Term
Cost Reduction. ISESCO Journal of Science and Technology, Volume 1/Number 1,
pp. 40-44.
Suruhanjaya Tenaga (Energy Commission), 2014. Peninsular Malaysia Electricity
Supply Industry Outlook 2014, Malaysia: Suruhanjaya Tenaga (Energy Commission),
Published Number: ST(P)14/09/2014.
Suruhanjaya Tenaga (Energy Commission), 2015. Statistics, Energy Supply. [Online]
Available at:
http://meih.st.gov.my/statistics?p_auth=F1SU9reQ&p_p_id=Eng_Statistic_WAR_S
TOASPublicPortlet&p_p_lifecycle=1&p_p_state=maximized&p_p_mode=view&p_
p_col_id=column-
1&p_p_col_pos=1&p_p_col_count=2&_Eng_Statistic_WAR_STOASPublicPortlet_
execution=e2s1&_Eng_Stat
Suruhanjaya Tenaga (Energy Commission), 2015. The Malaysian Grid Code,
Malaysia: Suruhanjaya Tenaga (Energy Commission).
Suruhanjaya Tenaga, 2012. EC- The Malaysian Distribution Code. Malaysia: s.n.
T. Aziz, M. H. T. S. N. M., 2013. VAR Planning With Tuning of STATCOM in a
DG Integrated Industrial System. Power Delivery, IEEE Transactions, Volume 28,
Issue:2, pp. 875-885.
117
T. Shigematsu, T. K. H. D. T. H., 6-10 October 2002. Applications of a vanadium
redox-flow battery to maintain power quality. Japan, IEEE. DOI:
10.1109/TDC.2002.1177625, pp. 1065-1070.
Tande, J. O. G., 2003. Grid Integration of Wind Farms, Wind Energy. John Wiley &
Sons, Ltd., pp. 281-295.
Teh, C., 2010. CHRISTOPHER TEH. [Online]
Available at: http://www.christopherteh.com/blog/2010/11/wind-energy/
[Accessed 9th September 2015].
Tejas Zaveri, B. B. N. Z., 2012. Comparison of control strategies for DSTATCOM in
three-phase, four-wire distribution system for power quality improvement under
various source voltage and load conditions. International Journal of Electrical
Power & Energy Systems,Volume 43, Issue 1, pp. 582-594.
TNB, D. A. H. A. S. S. D. I. Z. F. b. H. M. B. A. M. B. P. D. I. A. A. S. M. A. P. I. D.
R. V. K. D. A. R., 2012. TNB Technical Guidebook on Grid-interconnection of
Photovoltaic Power Generation System to LV and MV Networks, Malaysia: Tenaga
Nasional Berhad.
Trichakis, P. et al., 2006. An Investigation of Voltage Unbalance in Low Voltage
Distribution Networks with High Levels of SSEG. Newcastle upon Tyne, IEEE, pp.
182-186.
Trichakis, P., Taylor, P., Lyons, P. & Hair, R., 2008. Predicting the technical impacts
of high levels of small-scale embedded generators on low-voltage networks. IET
Renewable Power Generation, pp. 249-262.
Union of Concerned Scientists, 2014. Environmental Impacts of Natural Gas.
[Online]
Available at: http://www.ucsusa.org/clean_energy/our-energy-choices/coal-and-
other-fossil-fuels/environmental-impacts-of-natural-gas.html#bf-toc-3
118
von Appen, J. et al., 2011. Assessment of the Economic Potential of Microgrids for
Reactive Power Supply. The Shilla Jeju, Korea, LBNL, p. 8.
Wilkins, B. L. a. A. J., 2014. Designing to Mitigate Effects of Flicker in LED
Lighting: Reducing risks to health and safety. IEEE, pp. 18-26.
Wong, J., 2015. Fuzzy Controlled Energy Storage System for Low-Voltage
Distribution Networks with Photovoltaic Systems, Malaysia: UTAR.
Wong, J., Lim, Y. S. & Morris, E., 2014. Novel Fuzzy Controlled Energy Storage for
Low-Voltage Distribution Networks with Photovoltaic Systems under Highly Cloudy
Conditions. ASCE Journal of Energy Engineering, p. 17.
Wong, J., Lim, Y. S., Taylor, P. & Morris, S., 2011. Optimal utilisation of small-
scale embedded generators in a developing country - A case study in Malaysia.
Renewable Energy, 36, p. 12.
Wu, H. & Liu, Y., 2011. Novel STATCOM Control Strategy for Wind Farm Reactive
Power Compensation. Wuhan, IEEE, pp. 1-5.
Xi, Z., Parkhideh, B. & Bhattacharya, S., 2008. Improving Distribution Syatem
Performance with Integrated STATCOM and Supercapacitor Energy Storage System.
Rhodes, IEEE, pp. 1390-1395.
Yong, J. Y., Ramachandaramurthy, V. K., Tan, K. M. & Mithulanantha, N., 2015.
Bi-directional electric vehicle fast charging station with novel reactive power
compensation for voltage regulation. Electrical Power and Energy Systems, Volume
64, pp. 300-310.
Yun, J. Y., Yu, G., Kook, K. S. & Chang, B. H., 2013. A Study on the Criteria for
Setting the Dynamic Control Mode of Battery Energy Storage System in Power
Systems. Transactions of the Korean Institute of Electrical Engineers, 62(4), pp.
444-450.
119
Zaharim, A., Najid, S. K., Razali, A. M. & Sopian, K., 2009. Wind Speed Analysis in
the East Coast of Malaysia. European Journal of Scientific Research;Jun2009, Vol.
32 Issue 2, p208, pp. 208-215.
Zito, R., 2010. Energy Storage A New Approach. Canada: Scrivener.
Zuhairuse Md Darus, N. A. H. S. N. A. M. M. A. A. R. K. N. A. M. O. A. K., 2009.
The Development of Hybrid Integrated Renewable Energy System (Wind and Solar)
for Sustainable Living at Perhentian Island, Malaysia. European Journal of Social
Sciences – Volume 9, Number 4 (2009), pp. 557-563.
120
Appendix A: Specifications of SMA Sunny Island 5048
Output data
Nominal AC voltage (adjustable) 230 V (202 V – 253 V)
Grid frequency adjustable 45 Hz – 55 Hz
Continuous AC output at 25 °C /
45 °C 5000 W / 4000 W
Continuous AC output at 25 °C for
30 / 5 / 1 min 6500 W / 7200 W / 8400 W
Nominal AC current 21 A
Max. current 100 A (for 100 ms)
Output voltage harmonic distortion
factor < 3 %
Power factor –1 to +1
Input data
Input voltage (range) 230 V (172.5 V – 250 V)
Input frequency 50 Hz (40 Hz – 70 Hz)
Max. AC input current (adjustable) 56 A (2 A – 56 A)
Max. input power 6.7 kW /12.8 kW
Battery data
Battery voltage (range) 48 V
Max. battery charging current 120 A
Continuous charging current 100 A
Battery capacity 100 Ah to 10000 Ah
Charge control IUoU with automatic full and equalization
charge
Efficiency/power consumption
Max. efficiency (typical) 95 %
Own consumption with no load
(standby) 25 W (< 4 W)
Protection type DIN EN 60529
Certification CE
Device protection short-circuit, overload, over temperature
Interfaces
2 LEDs, 4 buttons, 2-line display, 2
multifunction relays, RS485 /
RS232 / CAN electrically separated (optional),
MMC/SD card
Mechanical data
Width / height / depth 18.4 / 24.1 / 9.3 inch (467 / 612 / 235 mm)
Weight 139 lbs (63 kg)
Ambient conditions
Ambient temperature –13°F ... 122°F (–25 °C to +50 °C)
Warranty 2 years
Accessories
Ext. battery temperature sensor Included
“GenMan” generator manager Optional
121
Appendix B: Program written in Data Taker DT82
begin "UTAR"
RA100-cv 'trigger when 100cv change to zero
1MODBUS(AD10,R4:1)=1142
RB100+CV 'trigger when 100cv change from zero
'Seq for modbus40001 to start motor
'a) xxxx x1xx x111 x110 (1142d)
'b) xxxx x1xx x111 x111 (1143d)
'c) xxxx x1xx x111 1111 (1151d)
'Write to Modbus slave (AD9 is big inverter, 10 is small inverter)
1MODBUS(AD10,R4:1)=1142
1MODBUS(AD10,R4:1)=1143
1MODBUS(AD10,R4:1)=1151
RC1S
'40002 contains set frequency
1MODBUS(AD10,R4:2)=101CV 'Inverter frequency
RD10S 'Read power meter values
1MODBUS(AD2,R3:29,=1..10CV)
delay=100
1MODBUS(AD2,R3:39,=11..20CV)
delay=100
1MODBUS(AD2,R3:49,=21..26CV)
delay=100
122
1MODBUS(AD2,R3:67,=27..34CV)
delay=100
1MODBUS(AD3,R3:29,=201..210CV)
delay=100
1MODBUS(AD3,R3:39,=211..220CV)
delay=100
1MODBUS(AD3,R3:49,=221..226CV)
delay=100
1MODBUS(AD3,R3:67,=227..234CV)
delay=100
1MODBUS(AD4,R3:29,=301..310CV)
delay=100
1MODBUS(AD4,R3:39,=311..320CV)
delay=100
1MODBUS(AD4,R3:49,=321..326CV)
delay=100
1MODBUS(AD4,R3:67,=327..334CV)
delay=100
1MODBUS(AD5,R3:29,=401..410CV)
delay=100
1MODBUS(AD5,R3:39,=411..420CV)
delay=100
123
1MODBUS(AD5,R3:49,=421..426CV)
delay=100
1MODBUS(AD5,R3:67,=427..434CV)
delay=100
1MODBUS(AD6,R3:29,=501..510CV)
delay=100
1MODBUS(AD6,R3:39,=511..520CV)
delay=100
1MODBUS(AD6,R3:49,=521..526CV)
delay=100
1MODBUS(AD6,R3:67,=527..534CV)
delay=100
1MODBUS(AD10,R4:2,=102CV)
END